Session FP1 - Poster Session III.
POSTER session, Tuesday morning, November 12
Grand Ballroom CDE, Rosen Centre Hotel

[FP1.001] Plasma Processing, Thrusters, Displays

[FP1.002] Large area beam-generated plasmas for materials processing applications

The Large Area Plasma Processing System (LAPPS) utilizes mA/cm^2, kilovolt electron beams confined by 100-200 gauss magnetic fields to produce large area, low temperature, high density plasmas in a variety of gases.[1] 1 cm thick sheet plasmas of up to 50 x 100 cm have been generated in pulsed and cw systems in proximity to grounded or biased platforms. Negative ion plasmas have also been generated during the beam pulse in attaching gases. A wide range of plasma and particle diagnostics have been fielded and results agree with analytic and particle code predictions. Initial processing tests have been performed on Si and photo-resist etching. Substrate bias voltage and working gas mixture dependencies have been measured. Applications of the technology to large area sputter coating have also been investigated. Progress in experiments and processing tests will be presented.

[1]R.A. Meger, et al, Phys. of Plasmas 8(5), p. 2558 (2001)

[FP1.003] Improved Plasma Isotope Enrichment using Magnetic Mirrors

A system of magnetic mirrors is used to improve the performance of a plasma isotope enrichment device. The mirror system provides an effective increase in the length of the device. This allows construction of a smaller device than previous devices. The magnetic mirrors also provide a new and improved method of discrimination between resonant particles and the non-resonant particles. This improved discrimination results in a higher level of enrichment, or a higher throughput rate for the device. Dawson and colleges first demonstrated that ICR heating could be used to enrich isotopes (Phys Rev Let, 37(23), 1976). The process relies on the mass dependence of the cyclotron frequency and couples energy into the cyclotron motion of a resonant ion species. As the ions drift across a uniform magnetic field region (0.6 m, 0.8T), the resonant ions gain large amounts of perpendicular energy. A carefully controlled shallow magnetic mirror is used to discrimination between the different ion species, allowing reflection of the resonant ions. Experiments have been conducted using a number of different isotopes (O, K, Rb, and Cs). Extensive experiments have been done with Rb, showing enrichment of Rb-87 from the natural 27% to over 96% in a single pass, using a small prototype device with water-cooled copper magnets. Work supported by DOE.

[FP1.004] Development of RF Plasma Torch at Atmospheric pressure

A RF torch operates by feeding helium or argon gas through two coaxial electrodes that are driven by a 13.56MHz radio frequency source of power ranging from 40W to 400W. In order to prevent an arc discharge, a dielectric material such as quartz is loaded outside of the center electrode that is connected to the RF power source. The inner surface of the outer-grounded electrode is also dielectric loaded. A stable, arc-free discharge was produced at a flow rate of 1.5 l/min of helium gas. Various reactors have been developed including water grounded electrode. According to the experimental data, the breakdown voltage varies for different diameters of inner electrode. An asymmetric biaxial reactor can ignite plasma at low breakdown voltage, which is less than that of symmetric coaxial reactor. However, plasma produced also asymmetrical distribution. The plasma characteristics are diagnosed, making use of optical emission spectroscopy (OES). The spectral lines are monitored according to various conditions of RF plasma torch. The lines to be analyzed are mostly in the visible and near infrared regions, and are observed through axial direction. Experimental results agree reasonably well with theoretical predictions.

[FP1.005] Reduction of Perfluorocompound Emissions by Microwave Plasma Torch

Surface cleanings are performed within a reduced pressure chamber by making use of perfluorocompounds (PFC) gases, which eventually contaminate the atmosphere. These contaminant gases are emitted with nitrogen gas, which is used for pump purges. In order to destruct all of the global warming gases including PFCs, we have developed a plasma abatement device, an electrodeless microwave plasma torch operated at the atmosphere pressure. The plasma abatement device is attached to the vacuum pump, which discharges the nitrogen gas with contaminants. The abatement was carried out using oxygen and air as an additive gases. The destruction and removal efficiency (DRE) of more than 98was achieved for tetrafluoromethane(CF4). The detailed characterization of CF4 abatement using Fourier Transform Infrared (FT-IR) and Quadrupole Mass Spectrometer (QMS) showed the major PFC by-products. Finally, experimental results indicate that the plasma abatement device for PFC destruction can be successfully used to abate all of the global warming gases in the semiconductor industry.

[FP1.006] Synthesis of Iron Nanoparticles in Plasma*

Previous results at Idaho have shown that Carbon-Iron nanoparticles can be synthesized by a plasma enhanced chemical vapor deposition process. The nanoparticles can be then be utilized to fill the channels of nanochannel glass thus forming a photonic crystal. We have constructed a new deposition chamber dedicated to understanding the role of plasma in nanoparticle formation. The channel glass is mounted on a custom-designed, differentially pumped sample holder and 10 Torr of differential pressure is applied across the 5 mm long nanochannels to drive the nanoparticles through the chanels. Precursor ferrocene ((C5H5)2-Fe) is the iron and carbon source. We will optimize the synthesis of these nanoparticles with reference to parameters such as rf power, background pressure, and sample location. A study of plasma parameters using Langmiur probes will be undertaken to measure the plasma temperature and density. Parametric effects on the synthesis of the particles will be studied. A Plasma Oscillation Probe (POP)1 will be installed as an alternative method for plasma characterization. Optimal nanoparticle growth conditions will be studied. Special attention will be given to the role of the plasma sheath in nanoparticle formation. Different plasma sources (rf and DC) will be compared. 1. Shirakawa and H. Sugai, Jpn. J. Appl. Phys. 32(1993)5129. *Work supported by Office of Naval Research and Idaho NSF EPSCoR funding.

[FP1.007] ECR plasma enhanced MOCVD system and the plasma role in film epitaxial growth of GaN and AlN

Abstract The development of low-dimension structure materials that are very promising for application of electronic device and optoelectronic device depend on the improvement of the technologies of epitaxial growth and characterization. The film growth technology with noninvasive, in situ, real time monitoring is becoming increasingly important as materials structure become more and more complex. A ECR plasma enhanced MOCVD (PEMOCVD) equipment (ESPU-U) with reflection high-energy electron diffraction (RHEED) for the first time has been developed. Multi-cusp Cavity-coupling ECR plasma source was adopted to provide reactive precursors in ESPD-U therefore the growth temperature was decreased and the working pressure was decreased down to the region less than 1Pa,which makes RHEED in situ monitoring possible. In this paper, the structure, key technologies and unique functions of ESPD-U will be introduced systematically. The epitaxial growth of the large lattice mismatch hetero-junction, GaN/ (0001) Al2O3 and AlN/ Al2O3 (0001), by PEMOCVD in the equipment with in situ RHEED monitoring and the important role of plasma in epitaxial growth were investigated. To remove the native oxidation layer producing a fresh substrate surface with atomic level flatness and to establish a template for the epitaxial growth H2(orH2/N2)-plasma cleaning in situ at 550-650¡æ for 2min- 20 min and N2 ¨Cplasma nitriding at 450-550¡æ for 1 min -30min after the cleaning for the surface of (0001) Al2O3 were investigated. Then the epitaxial growth was started by a two-step process including GaN buffer layer growth at low temperature ~500¡æ- 650¡æ for ~20nm and subsequent film growth of GaN and AlN at high temperature ~650¡æ-700¡æ. The films of GaN and AlN were characterized by the RHEED, XRD, AFM. The FWHM of GaN (0002) diffraction peak from 0.5amp;micro;m thick GaN film was 18 min and the FWHM of AlN (0002) diffraction peak from 0.3amp;micro;m thick AlN film was 12 min. The details of the work will be reported at the conference.

[FP1.008] Optimization of Ablation Plasma Ion Implantation

A new, optimized configuration for (laser) Ablation Plasma Ion Implantation (APII)[1] has been invented (target perpendicular to substrate), for which we have demonstrated: a) pure ion implantation with minimal contamination, b) elimination of arcing when the bias voltage is applied simultaneously with the laser pulse, c) higher ion dose by a factor of 2, and d) full energy ions by acceleration on only the voltage flattop. A theory of implanted ion current has been developed, with a novel scaling law, which agrees with the experimental results on microsecond time scales.

[1] B. Qi, R.M. Gilgenbach, Y.Y. Lau, M.D. Johnston, J. Lian, L.M. Wang, G. L. Doll and A. Lazarides, APL, 78, 3785 (2001)

Research supported by NSF.

[FP1.009] Platinum Coating on Tungsten Electrodes via Magnetron Sputtering

Various magnetron sputtering techniques are exploited in order to make bio-compatible Pt coating on W and some other metals. The substrate metals are mostly in the 3D shape, and some are cylindrical in the thickness of less than 0.5 mm. Platinum deposits are in the order of 10 to 100 nm. Biased and bipolar sputterings are tried with some microwave power applied. Nonlinear diffusion wave strokes in the surface medium will be discussed.

*Project supported by Ministry of Commerce, Industry, and Energy, Korea

[FP1.010] Density Gradient Dependent Helicon Modes

Radially localized helicon modes have been proposed to provide a fuller description of helicon discharges over a wide span of operating conditions and gas types [1]. These plasma modes could be of vital importance to the VASIMR engine. They depend on a radial density gradient and appear to operate over a range of frequencies inaccessible to traditional helicon discharges. Our work focuses on confirming experimentally the existence and properties of these helicon modes in Argon, Helium, and Hydrogen. We investigate the density profile, power deposition, wavefields, and dispersion relation of the new helicon modes which differ substantially from the properties of the traditional helicon plasma. We are using a set of dual half-turn helical antennas driven at 13.56 MHz. Our diagnostics includes a system for monitoring the plasma impedance, a set of Langmuir probes, a set of magnetic probes, as well as sensors for monitoring the pressure and DC magnetic field.

*Work supported in part by Advanced Space Propulsion Lab, Johnson Space Center, NASA

[1] B. N. Breizman and A. V. Arefiev, Phys. Rev. 84, 3863 (2000)

[FP1.011] MHD Simulation of Magnetic Nozzle Plasma with the NIMROD Code: Applications to the VASIMR Advanced Space Propulsion Concept

A simulation study with the NIMROD code [1] is being carried on to investigate the efficiency of the thrust generation process and the properties of the plasma detachment in a magnetic nozzle. In the simulation, hot plasma is injected in the magnetic nozzle, modeled as a 2D, axi-symmetric domain. NIMROD has two-fluid, 3D capabilities but the present runs are being conducted within the MHD, 2D approximation. As the plasma travels through the magnetic field, part of its thermal energy is converted into longitudinal kinetic energy, along the axis of the nozzle. The plasma eventually detaches from the magnetic field at a certain distance from the nozzle throat where the kinetic energy becomes larger than the magnetic energy. Preliminary NIMROD 2D runs have been benchmarked with a particle trajectory code showing satisfactory results [2]. Further testing is here reported with the emphasis on the analysis of the diffusion rate across the field lines and of the overall nozzle efficiency. These simulation runs are specifically designed for obtaining comparisons with laboratory measurements of the VASIMR experiment, by looking at the evolution of the radial plasma density and temperature profiles in the nozzle. VASIMR (Variable Specific Impulse Magnetoplasma Rocket, [3]) is an advanced space propulsion concept currently under experimental development at the Advanced Space Propulsion Laboratory, NASA Johnson Space Center. A plasma (typically ionized Hydrogen or Helium) is generated by a RF (Helicon) discharge and heated by an Ion Cyclotron Resonance Heating antenna. The heated plasma is then guided into a magnetic nozzle to convert the thermal plasma energy into effective thrust. The VASIMR system has no electrodes and a solenoidal magnetic field produced by an asymmetric mirror configuration ensures magnetic insulation of the plasma from the material surfaces. By powering the plasma source and the heating antenna at different levels it is possible to vary smoothly of the thrust-to-specific impulse ratio while maintaining maximum power utilization.

[1] http://www.nimrodteam.org [2] A. V. Ilin et al., Proc. 40th AIAA Aerospace Sciences Meeting, Reno, NV, Jan. 2002 [3] F. R. Chang-Diaz, Scientific American, p. 90, Nov. 2000

[FP1.012] Scaling of the VASIMR thruster first stage operation

An effective helicon plasma source [1,2] is used in the variable high specific impulse VASIMR plasma thruster [3]. Experimental prototypes – VX-3 and recently up-scaled VX-10 [4] configurations operate with hydrogen, deuterium and helium plasmas. A set of models [5-7] has been developed to study VASIMR light gases helicon discharge. Using zero-dimensional model incorporating energy and mass balance equations we study scaling of the plasma source efficiency with the increased mass flow rate, applied electrical power and dimensions of the quartz tube. We compare theoretical results with existing experimental data.

[1] M.A.Lieberman, A.J.Lihtenberg, 'Principles of ..', Wiley, 1994; [2] F.F.Chen, Plas. Phys. Contr. Fus. 33, 339, 1991; [3] F.Chang-Diaz et al, Bull. APS 45 (7) 129, 2000; [4] J.Squire et al., Bull. APS 45 (7) 130, 2000; [5] O.Batishchev, K.Molvig, AIAA technical paper 2000-3754, 2001; [6] O.Batishchev, K.Molvig, IEPC-01-208 paper, 27th Int. Electrical Propulsion Conf., 2001; [7] O.Batishchev, K.Molvig, AIAA technical paper 2002-0347, 2002.

[FP1.013] Study of electron and ion transport in a Hall effect thruster

Plasma transport in the Hall effect thrusters was a subject of many studies [1-2]. Despite this fact, the origin of a so-called anomalous transport is not understood to this date. Therefore, theoretical and numerical models [3-4] assume ad-hoc cross-field diffusion coefficients, which may differ by several times from the classical Bohm result. To study the transport phenomenon we have developed several models. One of the models is 2-dimensional in space (for axial and azimuthal directions), and 3-dimensional in velocity. A similar geometry was adopted in works [5], but we try to push the simulation to the realistic scale (several centimeters), while keeping the minimum spatial resolution on the order of the Debye length. It is hoped that the numerical results will provide a better understanding of the anomalous transport in Hall thrusters due to the collective modes.

[1] M.Hirakawa, IEPC-97-021 technical paper, 1997; [2] V. I. Baranov et al., IEPC-95-44, Moscow, 1995; [3] J.Szabo et al, IEPC-01-341, Pasadena, 2001; [4] V.Blateau et al, AIAA technical paper 2002-4105, 2002; [5] M.Hirakawa and Y.Arakawa, IEPC-95-164 technical paper, 1995; AIAA-96-3195 technical paper, 1996.

[FP1.014] Control of plasma flow in Hall thruster with active and passive segmented electrodes

Segmented electrodes, made from metal or ceramic spacers and having sizes comparable to, but smaller than, the acceleration region in Hall thrusters, are shown to significantly affect the plasma potential distribution. Associated with the potential changes are changes in the electron temperature and electron transport. We show that these changes are connected to changes in the physical properties of the electrodes, such as secondary electron emission and conductivity. The plasma potential distribution depends also on the operating conditions of the thruster, as well as the geometry and precise placement of the electrodes relative to the magnetic field distribution. These results can be explained, in part, through a hydrodynamic description of the thruster.

This work was supported by the New Jersey State Commission on Science and Technology and by the US DOE under Contract No. DE-AC02-76CHO3073.

[FP1.015] ENHANCED IONIZATION IN THE CYLINDRICAL HALL THRUSTER

Conventional annular Hall thrusters do not scale efficiently to low power. An alternative approach, a 2.6 cm miniaturized cylindrical Hall thruster with a cusp-type magnetic field distribution, was developed and studied. Its performance was compared to that of a conventional annular thruster of the same dimensions. The cylindrical thruster exhibits discharge characteristics similar to those of the annular thruster but has much higher propellant ionization efficiency. Significantly, a large fraction of multicharged xenon ions might be present in the outgoing ion flux generated by the cylindrical thruster. The operation of the cylindrical thruster is quieter than that of the annular thruster. The characteristic peak in the discharge current fluctuation spectrum at 50-60 kHz appears to be due to ionization instabilities. In the power range 50-300 W, the cylindrical and annular thrusters have comparable efficiencies (h=15-32configuration, the voltage less than 200 V was not sufficient to sustain the discharge at low propellant flow rates. The cylindrical thruster can operate at voltages lower than 200V, which suggests that a cylindrical thruster might be designed to operate at even smaller power.

This work was supported by grants of AFOSR and DARPA

[FP1.016] Anode Sheath in Hall Thrusters

Steady-state operation of a Hall Thruster at moderate discharge voltages requires the presence of a negative anode sheath and a back-ion flow in order to limit the electron flux towards the anode. An increase of the discharge voltage leads to the increase of the discharge current; however, the electron temperature is limited by the secondary electron emission from the channel walls. Therefore at large discharge voltages the electron drift velocity at the anode may become on the order of or larger than the electron thermal velocity, rendering the sheath unnecessary or making it positive. A quasi one-dimensional steady-state model of the Hall thruster with a given temperature profile shows that the applied discharge voltage determines the operating regime: for discharge voltages greater than a certain value, the negative anode sheath and the back-ion flow disappear. Both “no sheath” and “positive sheath” regimes may be responsible for limitations on outgoing jet velocity and accelerating efficiency observed experimentally for Hall thrusters operated at high discharge voltages. This work was supported by the US DOE under contract No. DE-AC02-76CH03073.

[FP1.017] High frequency instabilities in Hall thrusters

Gradient-driven Rayleigh-type instabilities in Hall plasma thrusters were analyzed using linearized two-fluid hydrodynamic equations. Necessary instability conditions and a general criterion for stability of azimuthally propagating perturbations were derived. For a simplified model of the axial distribution of parameters inside the thruster channel, the growth rate of an unstable wave, resonant with the azimuthal electron flow, was obtained. To study experimentally the high-frequency (1-100MHz) phenomena in Hall thruster plasma a novel diagnostic setup, consisting of single Langmuir probe, special shielded probe connector-positioner, and electronic impedance-matching circuit, was successfully built and calibrated. Through simultaneous probing of the Hall thruster plasma at multiple locations, azimuthal high-frequency waves have been successfully identified and characterized. The frequency and phase relations of the detected waves are in good qualitative agreement with the theoretical predictions for the Rayleigh-type instability.

[FP1.018] Modeling Electron Transport in Two Dimensional Simulations of Hall Thrusters

A two dimensional (r,z) hybrid (fluid electrons, PIC ions and neutrals) code has been written in order to understand Hall thruster dynamics. At issue is the role played by the electron transport in these discharges. Previous simulations have used various models for the electron conductivity, from collisional or "classical" to Bohm. Experimentally, it has been observed that such conductivity is a complex function of axial distance. At high voltage, the electron transport is reduced (as compared to the Bohm estimate) in the acceleration region where the azimuthal electron flow is strongly sheared, concomitant with a reduction of the relative level of fluctuations. In our simulations, we use an empirical model for the conductivity based on these observations and apply it to two thruster geometries. The results are then compared with previous simulations.

[FP1.019] The effects of ionization and catalytic recombination on Hall discharge stability

Coaxial Hall-effect accelerators are low-pressure, cross-field discharges which are primarily used as highly efficient ion rockets. Hall discharges exhibit rich oscillatory behavior that is thought to be important to enhanced cross-field electron transport. This study expands a two-fluid, four-equation linear stability analysis performed by Morozov in 1972 (\emphZhurnal Tekhnicheskoi Fiziki \textbf42, 612-19) to include new physical effects. In the original analysis, local stability is dictated by gradients in the plasma density and magnetic field. Including a source/sink term to account for ionization and recombination changes the stability limits such that the discharge is unconditionally unstable. When the plasma is primarily recombining, oscillations are strongly damped at most angles of propagation. Since electron loss is dominated by catalytic recombination at the channel wall, this damping couples the plasma oscillations to the electron-wall interaction in the discharge. Results of the stability analysis are compared qualitatively to recent experimental investigations of a laboratory Hall discharge with alumina walls.

[FP1.020] Characterization of ionization instabilities in a Hall-effect accelerator

Hall-effect accelerators are low-pressure, cross-field discharges, which are primarily used as highly efficient ion rockets. Many aspects of the electron dynamics within the discharge of such an accelerator remain to be clarified, and in particular the role that instabilities and plasma-wall interactions can play in the diffusion towards the anode. The dispersion properties of low-frequency (below 200kHz) waves in a Hall-effect annular ion accelerator are presented. Two types of materials with different secondary electron emission properties were used as the discharge chamber wall: boron nitride and alumina. The discharge was operated mainly in the ionization branch of the current-voltage characteristic . Fluctuations in the discharge current and of the plasma density were simultaneously recorded, at three different positions, by electrostatic probes negatively biased to the ion current saturation branch. The general features and possible origin of the observed instabilities are discussed.

[FP1.021] Hall Thruster Plume Studies using the BeamServer Antenna Pattern Code

BeamServer is a ray tracing code developed to study the effect of plasma thruster plumes on satellite communication signals. Rays are launched from the antenna feed, traced through the region containing antenna reflectors and plasma, and terminated on an exit surface. The electric field on the exit surface is then used to calculate the far-field antenna pattern, using the radiation integral. To verify both the code operation and our thruster plasma density models a "ground test" is planned, where a Hall thruster will be operated in a vacuum tank and a microwave system will transmit through the plume. Direct comparisons with ground test experimental data can be made on the BeamServer exit plane. To facilitate this extensive studies of the electric field magnitude and phase as a function of frequency and spatial location have been made. In addition, the Hall thruster exhibits large amplitude plasma instabilities, typically in the 25 KHz range. We have added instability models to the BeamServer code, and added time as a variable. Fourier transforms can be applied to either the exit plane or far-field patterns. We will present these studies and discuss the planned ground test.

[FP1.022] Hall Thruster Impact Analysis on Digital Satellite Communications

Hall Thrusters will be used for stationkeeping of geosynchronous communications satellites. These thrusters produce an inhomogeneous plasma plume which may interfere with communications signals (1-20 GHz) to and from the satellite. The plume's effects include beam-pointing error, beam attenuation, and spectral modulation [1]. Many systems digitally modulate an RF carrier and use coherent detection methods. The plasma may degrade the performance of these systems by diminishing the SNR at the receiver by squinting the beam away and degrading the synchronization by phase modulating the signal. We simulate a receiver whose input signals have been corrupted by the plasma plume. These modified input signals are formed by using the results of a previously developed ray tracing code [1]. We will present performance measures including bit-error rates and link availability.

* Work supported by Lockheed Martin Corp. and the Texas Higher Education Coordinating Board.

[1] Hallock et al, J. Spacecraft and Rockets V. 39 No 1 pp 115-124

[FP1.023] Ion Drift and Ion Temperature in the MNX Nozzle Region

A novel diode laser based LIF diagnostic is used to measure argon ion drift velocity and ion temperature near the nozzle of the MNX (Magnetic Nozzle Experiment) device. The laser is injected along the magnetic field axis and the fluorescence of both Zeeman sigma components are detected. The fluorescence linewidth is a convolution resulting from mostly Doppler and Zeeman broadenings. By using a de-convolution technique, the resulting fluorescence signal yields both the ion temperature and the ion drift. Ion temperatures in the nozzle region are within the 0.05 to 0.25 eV range, which is comparable to what was observed in the MNX inner plasmas. Preliminary result suggests that, under specific plasma conditions, drifting ion populations are present within (or beside) the thermal ion distributions. Specifically, LIF measurements near the nozzle (10 cm) reveal the presence of a hot tail within the ion distribution function. Further away from the nozzle (22 cm) two ion distributions of comparable intensities are observed. The first one is unshifted and thus stationary, while the second is moving away from the nozzle with a 4 eV (6.5 GHz) drift. The relative intensity of the drifting distribution is important when the pressure near the nozzle remains low but quickly diminish with increasing pressure. The magnitude of the drift also significantly decreases with raising pressure. Colder ions (0.05 eV) are commonly observed under high pressure conditions.

[FP1.024] Plasma Source and Equatorial Characteristic Dependencies of The Mini-Magnetospheric Plasma Propulsion (M2P2) Prototype during Large Chamber Experiments

Magnetospheric Plasma Propulsion (M2P2) seeks the creation of a magnetic wall or bubble (i.e. a magnetosphere) that will intercept the solar wind and thereby provide high-speed propulsion with efficient propellant utilization and power requirements. For successful inflation of the magnetic bubble a beta of unity must be achieved along the imposed dipole field. This is dependent on the plasma parameters that can be achieved with plasma sources that provide continuous operation at the desired power levels. Investigations of a 1.5 cm radius Helicon plasma source to generate plasma along an imposed dipole magnetic field have been conducted in a new 3800 liter vacuum chamber at the University of Washington. Steady state plasma generation has been accomplished while maintaining a very low neutral gas pressure (< 10-5 Torr) in the chamber allowing for magnetized plasma to be inserted along the dipole field. Measurements of plasma characteristics have been made using swept asymmetric double Langmuir probes for several different experimental configurations. Results show a building of plasma density in the outer equatorial region of the dipole with a peaked profile that is highly dependent on source and dipole magnet geometry.

[FP1.025] Data Voltage Effects on Sustaining Discharge and its Implication on Luminescence in Plasma Display Panels

The sustain discharge in widely used AC-PDP cells with a three electrode system occurs in the space between the parallel-sustain electrodes of X and Y on the front glass plate. But the wall charge is formed on the dielectric layer covering the X and Y electrodes, and also formed on the dielectric layer over the address electrode Z on the rear glass plate. Data voltage is applied to the address electrode. The luminescence of the plasma display panel (PDP) under a certain wall charge configuration decreases, because the wall charge prevents the normal sustaining discharge. The wall charge accumulation on the rear glass plate can be experimentally adjusted by applying the data voltage to the address electrode during the sustaining discharge period. Influence of the data voltage on the discharge characteristics in cells of AC-PDP and on the luminescence of the surface discharge is investigated. It is found from experimental observation that the wall charge on the rear dielectric layer is not formed, whenever the applied data voltage during the sustaining discharge is 40~60 achieving the highest luminescence in AC-PDP.

[FP1.026] Measurement of Electron Temperature and Density in AC Plasma Display Panels

The present AC-PDP has several technical problems, including low brightness and low efficiency. In order to overcome these difficulties, it is necessary to find the fundamental properties of the plasmas generated from the electrical discharge in PDP cells. For the investigation of the basic parameters in AC-PDP plasma, the micro Langmuir probe and ICCD (Intensified Charge Coupled Device) camera have been used to diagnose electron temperature and plasma density in AC-PDP cells. It is observed from the experiment that the electron temperature, obtained from both the micro Langmuir probe and high speed camera, decreases from 1.8 eV to 0.9 eV as the pressure of the neon gas increases from 150 Torr to 350 Torr. It is noted that the measured electron temperatures from the Langmuir probe and high speed camera are in good agreement with each other within a 5£¥ error margin. The plasma density at the lateral distance of 125 §­ from the center of the sustaining electrode gap has been found to decrease from 3.7 ¢¥ 1011 cm-3 to 2.3 ¢¥ 1011 cm-3, as the gas pressure increases from 150 Torr to 350 Torr.

[FP1.027] MST and Reversed Field Pinch

[FP1.028] MST Progress and Plans

New diagnostic capability combined with turbulence control is producing a clearer picture of magnetic fluctuation-induced transport. Local stochastic transport depends on both the mode amplitudes and the density of nearby resonant surfaces. Radially resolved measurements of the electron heat diffusivity agree with theoretical estimates \sim v_th(b/B)^2, but only where the resonant surface density is high. With auxiliary current drive, broadband mode suppression reduces the local transport to create RFP global confinement comparable to tokamak plasmas. In this case an increased population of high energy electrons (inferred from hard x-ray emission) is inconsistent with the velocity dependence characteristic of stochastic diffusion. This plus direct measurement of reduced fluctuations in the core by FIR polarimetry evidence closed magnetic surface formation. Core electrostatic fluctuations are also measured with a heavy ion beam probe for the first time in an RFP. The nonlinear evolution of the magnetic spectrum is studied using bi-spectral techniques, and, for the first time, m=0 modes are measured to be linearly stable (in standard RFP plasmas). An oscillating BT, applied to test ac helicity injection, entrains sawtooth magnetic relaxation processes. Ion temperature profiles are obtained using CHERS and Rutherford scattering. For future plasma control, a 2nd generation lower hybrid antenna shows improved performance, and electron Bernstein wave coupling is high in improved confinement plasmas, self-consistent with profile control requirements. Also, a new pellet injector provides core fueling. Work supported by US DOE.

[FP1.029] Dynamics of Perpendicular and Parallel Plasma Momenta in the MST Reversed Field Pinch

The dynamics of the perpendicular plasma momentum and parallel plasma momentum over sawtooth crashes and associated magnetic reconnections is studied in the core and at the edge of reversed field pinch plasmas. Both the parallel and perpendicular plasma velocities exhibit sudden changes during the reconnections. The core plasma velocity is measured with Doppler spectroscopy and the edge velocity is measured with a Mach probe. The study reveals that the plasma parallel momentum relaxes over the plasma radius similar to the current profile relaxation predicted by Taylor's theory and experimentally observed in MST. The flattening of the parallel plasma momentum profile is in good agreement with the theoretical predictions of two-fluid MHD. Local measurements of the j\timesB plasma torque associated with magnetic fluctuations are done at the plasma edge with an insertable magnetic probe. The measurements indicate that the torque is much higher than the rate of change of the corresponding plasma momentum.

[FP1.030] Tokamak-like energy confinement with high beta in the MST RFP

Recent improvements in magnetic fluctuation control in the MST have led to a 10-fold increase in the energy confinement time, which reaches 10 ms. Concurrently, beta-total exceeds 15 percent, and beta-poloidal exceeds 18 percent. In addition to exceeding for the first time the historical RFP constant-beta confinement scaling, the 10 ms confinement time also falls within a factor of two of the predictions from the tokamak (ITER) L-mode and ELMing-H-mode scaling laws. The proximity of MST confinement to tokamak-scaling predictions may not be coincidental. High-energy runaway electrons up to 100 keV are for the first time confined in the MST plasma core, and diffusion of these electrons is independent of their velocity. This suggests a transition from the usual stochastic magnetic field to well-formed magnetic flux surfaces and that core transport may now be dominated not by magnetic fluctuations, but by electrostatic fluctuations instead. Work supported by USDOE.

[FP1.031] Hard X-ray Diagnosis of the Fast Electron Distribution in the MST RFP

Hard x-rays (5-100 keV) are recorded in MST only during improved confinement discharges produced by inductive current profile control. The emergence of hard x-rays indicates good confinement of runaway electrons in an RFP. Fokker-Planck modeling of the x-ray emission shows fast electron diffusion is small and independent of energy, evidence that transport is no longer dominated by magnetic fluctuations during the improved confinement period. The hard x-ray diagnostic has recently been upgraded to a 16-channel CdZnTe array. Rather than using conventional pulse height analysis, the shaped output pulses from each detector are directly digitized. This cost-effective technique allows for excellent energy resolution, dynamic time binning, and better noise rejection/pile-up detection. This array will enable profile measurements of fast electron production and transport, and will provide critical diagnostic support to RF current drive experiments on MST.

[FP1.032] Ion Temperatures and Magnetic Fluctuations in the Madison Symmetric Torus in Steady-State

Previous work (Scime et al., Phys Fluids B \textbf4 (12), 4062) has established a link between ion heating and magnetic fluctuations on MST during a Magnetic Reconnection Event (MRE). Here we report on observations in the steady-state, more than 0.2 ms before or after and MRE. In standard MST discharges, the steady-state ratio T_i/T_e > 0.5. This T_i is too large to be explained by a balance between classical collisional heating by electrons (P_ei) and charge exchange loss (P_CX). In MST discharges with Pulsed-Poloidal Current Drive (PPCD), magnetic fluctuations drop significantly, T_e increases by a factor of 2, P_ei increases by a factor of 5, the global energy confinement time \tau _e increases by a factor of at least 5, and P_CX decreases by a factor of 2. Nonetheless, as a result, T_i increases by only about 20% when compared to standard discharges, and can be explained by a balance between P_ei and P_CX. Magnetic fluctuations are measured by coils and flux loops in the plasma edge. The temperature of the bulk majority ion species is measured by Rutherford Scattering, while the temperature of fully-stripped impurity species is measured by CHERS; these temperatures are found to be equal. Work supported by US DOE.

[FP1.033] Modeling and measurements of magnetic stochasticity and transport in the MST reversed field pinch

Results from experimental measurements and modeling of stochastic transport are beginning to agree. The modeling was done with a field line tracing code, which uses spatial profiles from a 3-D nonlinear MHD code, DEBS, and incorporates experimentally measured edge fluctuations. The modeling finds good agreement with Rechester - Rosenbluth diffusion just inside the reversal surface, where the islands highly overlap, and diverges from Rechester - Rosenbluth elsewhere. Electron heat transport is measured experimentally and agrees with the model to within a factor of three. Measurements of the drop in electron thermal transport in plasmas where fluctuations are suppressed by current profile manipulation are in agreement with the model. Recent measurements of the confinement of run-away electrons observed in the core suggest reformation of magnetic flux surfaces from a previously stochastic field. The modeling clearly shows this transition in profiles of radial magnetic field line diffusion. In addition we have started modeling of fast ion motion and new results on the effect of magnetic field stochasticity on the ion confinement will be reported.

Work supported by U.S. D.O.E.

[FP1.034] Electron Bernstein Wave Heating in the Madison Symmetric Torus Reversed Field Pinch

A system to heat electrons in the Madison Symmetric Torus through the electron Bernstein wave is currently being developed. This is an attractive heating scheme for the overdense RFP plasma, where electron cyclotron heating and current drive are inaccessible. Low power experiments (\sim 1 watt) have shown that a significant fraction of launched electromagnetic power successfully couples to the electron Bernstein wave. Furthermore, these experiments have found an optimized launch with finite N_\perp. We present the initial results from a moderate power (\sim 150 kW), several millisecond experiment driven by a pair of S-band traveling wave tube amplifiers. Electromagnetic power will be injected into plasmas with similar Ohmic heating levels (200 - 300 kW), and the effects on the electron distribution will be monitored with x-ray detectors. This work is supported by USDOE.

[FP1.035] EBW Coupling Experiment in Overdense RFP Plasma

Experimental observation of blackbody emission from MST establish the viability of EBW heating in over-dense plasmas. However, a major technical challenge is to develop a robust technique for coupling high power to the EBW from external antennas. Our experimental results show the possiblities to couple energy to the EBW in RFP geometries. The antenna, two S-band waveguide with one common E-plane wall, launches wave with a finite N_\perp with an externally imposed phase shift between the waves in each waveguide. Power from a single TWT amplifier ( < 10 watts) at 3.6 GHz was divided and fed into two arms of twin waveguide through isolators and bidirectional couplers. A mechanical phase shifter was used to manually vary the interwaveguide phase between plasma shots. The amplitudes and phases of the forward and reflected powers were measured for each waveguide. Vacuum measurements agree well with theory. Coupling has also been measured in PPCD plasmas. During PPCD a substantial reduction (from 80% to 20%) in the reflected power was observed. The asymmetry in coupling with respect to N_\perp is clearly observed in measurements.

[FP1.036] Coupling to the Electron Bernstein Wave With Wave- guide Antennas: Theory and Experimental Results from MST

The electron Bernstein wave (EBW) is of interest for both diagnostic applications and for heating and current drive in low field devices such as present spherical torus experiments and the reversed-field pinch (RFP). In these devices, neither X- or O-modes can propagate in the interior of the plasma. We compare the predictions of a generalized waveguide coupling code~[1] to the experimental results from the MST RFP in which a pair of S-band waveguides were oriented to excite the X-mode (which couples to the EBW near the upper hybrid resonance) in the edge of the plasma. Good qualitative agreement between the predicted phase dependence of the reflection coefficient and the measured results is obtained. In particular, the predicted strong dependence of the coupling on the sign of the toroidal phasing was observed.\par \vskip3pt [1]~Pinsker, R.I., et al., in Radio Frequency Power in Plasmas (Proc.\ 14th Top.\ Conf., Oxnard, CA, 2001), (AIP, Melville, NY, 2001) p.~350.

[FP1.037] Interdigital Line Antennas for Launching LH Waves in MST

RF current drive has been proposed as a method for reducing the tearing fluctuations that are responsible for anomalous energy transport in the RFP. A system for launching lower hybrid slow waves at 800 MHz and n_||= 7.5 is now in operation on MST. The antenna is an enclosed interdigital line using \lambda/4 resonators with an opening in the cavity through which the wave is coupled to the plasma. A new antenna has been built incorporating several design improvements. These include larger vacuum feedthroughs, better impedance matching, internal instrumentation and improved directionality. Power handling is 3-4 times that of the original and continues to improve with conditioning. Further design improvements are underway to optimize impedance matching and damping rate along the traveling wave structure.

[FP1.038] Lower Hybrid Experiments in MST

Current drive using RF waves has been proposed as a means to reduce the tearing fluctuations responsible for anomalous energy transport in the RFP. A lower hybrid antenna that operates at 800 MHz is being used in MST to assess the feasibility of this approach. Antenna performance is affected by plasma sawteeth indicating that the antenna and the plasma are well coupled. The antenna has been instrumented with pickup loops in the backplane to measure power flow. The measured power damping length of approximately 1-2 wavelengths is consistent with the amount of launched power that travels completely through the antenna structure. The damping length increases with decreasing plasma density and is dependent on wave launch direction. Probes capable of measuring RF fields will be used to locate the lower hybrid wave in the plasma. In addition, soft and hard x-ray diagnostics will be used to look for evidence of RF-plasma interactions. Modelling with GENRAY and CQL3D of plasma-lower hybrid wave interactions is ongoing.

[FP1.039] Plasma Response to Oscillating Poloidal and Toroidal Current Drive in the Madison Symmetric Torus

Two 1 MVA 500 Hz oscillators have been installed in the toroidal and poloidal magnetic field circuits of the Madison Symmetric Torus (MST) Reversed Field Pinch. These devices drive alternating poloidal and toroidal currents in the edge of the plasma affecting the current profile and thus both the spectrum of the tearing mode fluctuations and energy confinement. We find the amplitude of the dominant (m=1) modes rises and falls with the oscillating edge poloidal current while the m=0 mode amplitude appears to do just the opposite. The ion temperature also oscillates with the edge poloidal current. The ion temperature excursion \tildeT_i / is as much as 40 percent. This pronounced oscillation does not appear in measurements of the electron temperature and has no classical explaination. These effects and an entrainment of the sawtooth instabilities with the applied oscillations are shown. These oscillators work together to test Oscillating Field Current Drive (OFCD). Low power partial current drive tests will be reported. Variables include mean plasma current and relative phasing of the two oscillators.

[FP1.040] Penetration of AC Fields into a Reversed Field Pinch in Oscillating Field Current Drive

AC magnetic fields are applied at the shell of the Madison Symmetric Torus (MST), and these penetrate into the plasma. The applied poloidal and toroidal fields are used in oscillating field current drive (OFCD). The initial OFCD system should be capable of driving a portion of the total plasma toroidal current. In order to study the effects of AC field propagation within MST, experiments are planned with several diagnostics, including motional Stark effect spectroscopy, far-infrared interferometry/polarimetry, Rutherford scattering, ion Doppler spectroscopy, charge exchange recombination spectroscopy, and Thomson scattering. In previous such experiments, the normally quasi-periodic sawtooth cycle was entrained to the OFCD cycle, and AC magnetic field penetration toward the core was detected during sawteeth. New experimental results are used in MSTFit equilibrium reconstruction and compared with MHD theory.

[FP1.041] Pellet Injection into MST RFP Plasmas

A four-barrel cryogenic pellet injector, designed and built by Oak Ridge National Laboratory, has been installed on MST. The injector is a pipe gun utilizing high-pressure hydrogen gas for acceleration of pellets. Presently, the two barrels in use accommodate deuterium pellets with diameters of 1.0 mm and 1.8 mm and lengths ranging from 1.5 mm to 2.7 mm which are injected radially into MST. Pellet speeds of 1300 m/s have been achieved in initial experiments, and many pellets cross the plasma diameter without completely ablating. The pellets rapidly increase the central density and peak the density profile, something not possible with gas puffing alone. Pellet injection into improved-confinement plasmas has allowed the achievement of line-averaged densities 10-20% larger than the usual limit, above which edge-resonant MHD instability is triggered, and confinement is degraded. Mechanical punches will soon be installed to allow slower pellet speeds.

Work supported by U.S.D.O.E.

[FP1.042] Spectral Motional Stark Effect Measurements in MST

Using the full measured spectrum of the Stark-split hydrogen emission from a diagnostic neutral beam we are able to measure magnetic fields as low as 0.2 T in MST. A fast liquid crystal shutter and CCD spectrometer are used to take 0.1 ms exposures of emission from a 30 keV hydrogen diagnostic neutral beam. Recent improvements will allow 7 independent measurements of the central toroidal magnetic field during a single shot, giving a strong constraint for equilibrium reconstructions. Key observations to date include a reduction of the toroidal field on axis during a sawtooth crash of about 10profile control, the central magnetic field is measured to increase slowly and during oscillating poloidal current drive, very little penetration of the oscillating edge magnetic field to the core is observed. Accuracy is determined in part by the validity of the assumption that the underlying components of the Stark manifold are statistically populated and emit with fixed relative intensities. These assumptions are being examined by detailed atomic modeling. Work supported by U.S.D.O.E.

[FP1.043] Heavy ion beam probe measurements of the equilibrium potential and electrostatic fluctuation profiles on the Madison Symmetric Torus

Measurements of the radial equilibrium potential profiles have been successfully obtained with a Heavy Ion Beam Probe (HIBP) in the core (0.25 < r/a < 0.75) of the Madison Symmetric Torus (MST) Reversed Field Pinch. Typically, \phi (r) has a magnitude of 1.0-2.0 kV in standard 380 kA discharges. The core profile of the electrostatic potential fluctuations \~\phi and electron density fluctuations \~n_e(r) have also been measured in MST. The measured \~\phi(r) ranges from 30-50 V_rms and \~n_e(r) ranges from 10-20% for this same standard 380 kA discharge. While most of the data obtained thus far have been for standard discharges at a variety of plasma currents, preliminary measurements have also been obtained for other discharge conditions, including biased discharges and pulsed poloidal current drive (PPCD) discharges. Confinement is significantly improved in PPCD discharges and HIBP measurements obtained thus far show changes in \phi (r), \~\phi(r) and \~n_e(r). The general status of HIBP measurements on MST will be presented including representative data from all types of discharges and measurement development issues.

[FP1.044] Density Fluctuation Measurements in the MST Reversed Field Pinch Using a Heavy Ion Beam Probe

Localized measurements of density fluctuations in the hot core of the Madison Symmetric Torus (MST) reversed field pinch have been made with a Heavy Ion Beam Probe (HIBP). The density fluctuations are 10 - 20% in the region 0.25<r/a<0.75. Results are presented for both standard and improved confinement discharges. To have confidence in the results it is necessary to account for possible signal corruption, such as ion path effects and electronic noise. The HIBP on MST has 2 additional factors that are not common to most HIBPs. The small machine ports (2" injection port and 4.5" detection port) can result in vignetting of the beam and the constantly changing magnetic fields causes the beam's sample location to move. It is shown that such instrument effects can be accounted for and are small in selected data sets.

[FP1.045] Core Magnetic Fluctuations and Current Profile Dynamics in the MST Reversed-Field Pinch

First measurement of the current density profile and magnetic field fluctuations in the core of a high-temperature reversed-field pinch are presented. We report three new results: (1) The current density profile is observed to peak during the slow ramp phase of the sawtooth cycle and flatten promptly at the crash. Measured core magnetic fluctuations are observed to increases four fold at the crash. The fluctuating current which generates the dominant large-scale magnetic fluctuations is measured and found to have a spatial extent ¡Ü8 cm about the rational surface. (2) The core magnetic fluctuations are observed to decrease four-fold in plasmas in which the energy confinement time improves ten-fold. (3) The parallel current density increases in the outer region of the plasma during PPCD. However, the current density also increases in the core, relative to standard plasmas. Current density peaking on axis may be explained by a reduction of the dynamo (anti) current drive in the core. Work is supported by U.S.DOE

[FP1.046] SXR imaging of chaos healing in the core of the Madison Symmetric Torus (MST) RFP

In the MST reversed-field pinch (RFP), experiments with current profile control (pulse parallel current drive, or PPCD) are routinely performed, allowing record values of temperature and confinement time for the RFP. Recently, a system of miniature soft-x-ray (SXR) cameras has been installed and operated, and an overview of the resultant tomographic data will be shown. Profiles of SXR emission during PPCD show clear oscillations at the poloidal rotation frequency of the innermost resonating tearing modes. An helical coherent structure is observed with QSH spectra. Moreover, we have observed the beating of two nearby frequencies in the x-ray data when all MHD modes have small amplitudes. This is consistent with the presence of two small islands in the plasma core. It is also evidence, reinforced by the results of the guiding center Monte Carlo code ORBIT, that the PPCD technique is capable of reducing magnetic chaos to the point that regions of closed magnetic flux surfaces are produced.

[FP1.047] Effects of fast particles on internal modes in reversed field pinches

Fast ion population in tokamaks is found to have a significant influence on the dynamics of global plasma modes. High energy particles appear to play an essential role in the suppression of sawtooth relaxation oscillations in tokamaks. These particles can also lead to the occurrence of fishbone-like instabilities. In a normal mode analysis of plasma oscillations the fast particles are described by drift kinetic equation in which the perturbed electric and magnetic fields are evaluated at the position of the particle's guiding center. In this treatment the effects due to spatial variation of perturbed fields within the gyroradius are not considered. In reversed field pinches (RFP) the magnetic field is an order of magnitude smaller than in tokamaks. The Larmor radius of neutral beam injected particles in the 10 KeV energy range in RFPs is a substantial portion of the minor radius. The perpendicular wave length of global plasma modes is comparable to the Larmor radius of fast particles; thus the finite Larmor radius effects mentioned above can be important. In this study we concentrate on these effects only. We apply a simple constitutive relation without drift contributions for the fast particle dielectric response obtained from the linearized Vlasov equation. We consider cylindrical RFP equilibrium and use the resistive MHD model for the description of the plasma bulk. The changes to the growth rates of the internal modes due to the fast particles with large gyroradius are calculated.

[FP1.048] Impurity ion measurements on MST

Measurement of impurity ion dynamics is essential for understanding anomalous ion heating, momentum relaxation, fluctuation induced particle transport and the MHD dynamo. Charge exchange recombination spectroscopy (CHERS) provides a localized measurement of the density, flow and temperature of fully-stripped impurity ions. Standard CHERS operation in MST uses C VI emission at 343.4 nm and is localized to +/- 1 cm. CHERS measurements with an existing Ion Doppler Spectrometer (IDS) yield ion temperatures close to electron temperatures, with a time resolution limited to about 1 ms. Impurity density profiles have also been measured with this diagnostic and are shown to have an unexpected hollow profile. A new Ion Doppler Spectrometer (IDS II) is currently being constructed and should increase time resolution to at least 100 ms, to better observe fluctuations. A Ross filtered spectrometer is also used to view transitions to the ground state and allows for more precise time resolved density measurements. Data of localized time resolved density measurements using the CHERS beam and the Ross filtered spectrometer will be presented. Atomic modeling to consider fine structure corrections to the CHERS line shape is being carried out and will also be presented.

Work supported by U.S.D.O.E.

[FP1.049] measurements of m=0 excitation and nonlinear coupling in MST

The generation of m=0 magnetic fluctuations in the reversed field pinch (RFP) is not fully understood. Large-scale magnetic fluctuations have been measured in the MST(Madison Symmetric Torus) using edge magnetic coils in a toroidal array. The ExB velocity is measured through a sawtooth cycle using Langmuir probes and decomposed using a pseudo-spectrum method. The measured quantities are combined to derive the terms in the MHD equations responsible for linear excitation of the m = 0 modes by the mean fields. Experimentally, the magnetic fluctuation of poloidal mode number m = 0 appears to be damped throughout the standard RFP sawtooth cycle. This agrees with the DEBS code. Three wave nonlinear mode coupling with m = 0 modes is measured through a sawtooth cycle via bi-spectral analysis. The result suggests that m = 0 excitation is driven by nonlinear mode coupling, also consistent with computation. An examination of m=0 excitation in non-standard plasma will be presented.

Work supported by U.S.D.O.E

[FP1.050] Technique development of spectroscopic ion beam imaging for mapping magnetic fields in a plasma

The trajectory of an ion beam as it passes through a magnetically confined plasma is determined by the ion mass, energy, charge state, and the magnetic field structure. In undergraduate physics labs, students use a measure of beam deflection in a well defined magnetic field to determine the charge-to-mass ratio of a particle. The complementary analysis is equally valid: given a known charge-to-mass ratio, the magnetic field may be determined. Additional complexity is introduced in a spatially non-uniform, time varying magnetic field, such as that found in a reversed field pinch plasma. The technique of field mapping via spectral imaging is being developed with a Heavy Ion Beam Probe on the Madison Symmetric Torus. Spectral measurements of emission lines of heavy beam ions, at multiple views displaced both poloidally and toroidally from the injection port, will be discussed. The addition of CCD cameras will enable a three-dimensional reconstruction of the beam trajectory through the plasma, which can be used to determine the components of the equilibrium magnetic field.

[FP1.051] Current Transport Simulations of the MST RFP with a Tearing-mode Based Hyper-resistive Coefficient.

Hyper-resistivity (a helicity-conserving current-diffusion term in Ohm's law)(A.\ H.\ Boozer, J. Plasma Pysics, 35 (part 1), 133 (1986).), is frequently used to model the effects of magnetic perturbations due to magnetic islands or low-amplitude turbulent fields on current transport across mean magnetic surfaces. This process is important in the magnetic-field evolution of toroidal devices such as the RFP and the edge-driven spheromak and can provide a mechanism for sustained operation. In previous work(H.\ L.\ Berk et al., Bull.\ Am. Phys.\ Soc., (Oct., 2000), paper VP1.55.), we presented a quasi-linear calculation of the hyper-resistivity coefficient in a cylinder due to tangled magnetic field-lines resulting from multiple tearing modes. In this paper, we present results of simulations of the MST RFP using this hyper-resistive coefficient in the 1 amp; 1/2 D transport code Corsica(http://wormhole.ucllnl.org/caltrans/). Stability at the m=1 resonant surfaces is obtained from cylindrical \Delta', and island widths are obtained from the Rutherford island equation; these are solved self-consistently with the evolving equilibria. We explore the resulting current profiles, in particular the steep cliff that forms at the reversal surface, and the study the dependence on parameters.

[FP1.052] Single and Multihelical States in Reversed Field Pinches

Single helicity states in reversed field pinches are predicted by 3D MHD simulations; similar states, called 'quasi-single helicity states', have been observed in several experiments. Here, we explore the behavior of these states to different dissipation rates, initital conditions, boundary conditions, and sheared toroidal rotation. Previous work [J. M. Finn, R. A. Nebel, C. G. Bathke, Phys. Fluids B, (1992)] has indicated that single helicity states occur if the dissipation is high enough. This result has been validated by our present work which has tracked these solutions for 10s of resistive times. Previous work has also indicated that these states persist in less dissipative plasmas provided that the proper initial conditions are imposed on the plasma. Results will be presented for long time (resistive time scale) simulations for these conditions as well. For toroidal rotation, a narrowing of the spectrum of m=1 modes is expected, based on the idea that nonlinear coupling of modes drives other modes but at a phase velocity different from the plasma velocity at the mode rational surface of the driven mode. We will show results for these cases also.

[FP1.053] Current profile control experiments in EXTRAP T2R

EXTRAP T2R is a high aspect ratio (R=1.24 m, a = 0.183 m) reversed-field pinch device, characterised by a double, thin shell system. The simultaneous presence of many m=1, |n| > 11 tearing modes is responsible for a magnetic field turbulence, which is believed to produce the rather high energy and particle transport that is observed in this type of magnetic configuration. In this paper first results from current profile control experiments (PPCD) in a thin shell device are shown. When an edge poloidal electric field is transiently applied, an increase of the electron temperature and of the electron density is seen, which is consistent with an increase of the thermal content of the plasma. At the same time, the soft x-ray emission, measured with a newly installed miniaturised camera, shows a peaking of the profile in the core. Furthermore, the amplitudes of the m=1 tearing modes are reduced and and the rotation velocities increase during PPCD, which is also consistent with a reduction of magnetic turbulence and a heating of the plasma

[FP1.054] Numerical analysis of magnetic topology in the Reversed Field Pinch

We discuss results concerning the application of ORBIT, a hamiltonian guiding centre Monte Carlo code with numerical or analytic magnetic equilibria, to Reversed Field Pinch configurations. In particular we have analyzed the structure of magnetic field during reduced magnetic chaos regimes obtained with Pulsed Poloidal Current Drive experiments and in Quasi Single Helicity conditions. The code has been used simulating operational conditions of the RFX and MST devices. Codes output are compared with experimental results, in particular with tomographic imaging of the plasma.

[FP1.055] Controling MHD mode dynamics in an RFP with rotating helical field

Three critical MHD issues are to be addressed in the course of development of a reversed field pinch (RFP) fusion reactor : internal tearing modes, internal and external ideal kink modes. In STE-2 RFP [R/a=0.4m/0.1m], efforts have been made to actively control the resistive wall tearing modes with external rotating helical field (RHF). Without RHF, core resonant mode (m=1/n=8) grows with time scale of \tau_w, and identified as the resistive wall tearing modes. The resonant mode, which otherwise is locked to the wall, rotate with the applied RHF with perturbation level of \sim0.4 %, and that the applied RHF has an influence on the rotation of the neighboring modes as well. When we increase the perturbation amplitude up to 0.6 %, the pinch parameter \Theta sometimes increases to higher than 2.5. The high-\Theta operation is the result of plasma current increase more than toroidal flux. In this high-\Theta regime, the resonant mode sometimes rotates in the direction of plasma current at the same velocity as the applied RHF, while the neighboring modes rotate at lower velocity. The time averaged fluctuation level of the dominant modes remains lower than in the case without RHF. One of the interesting features in high-\Theta regime RFP is that m=1/|n|=3,4 modes grow with the time scale of \tau_w. The behavior will be discussed in comparison with linear stability of external kink modes.

[FP1.056] Effect of multiple resistive shells on the dynamics of the slinky mode in RFP plasmas

The effects of multiple resistive shells and transient electromagnetic torque on the dynamics of mode locking in the RFP plasmas are studied. We study the EM torque acting on the tearing modes produced by a system of resistive shells. These shells may consist of several nested thin shells or several thin shells enclosed within a thick shell. Both the steady state theory and the time dependent theory are developed. The steady state theory is shown to provide accurate account of the resultant EM torque if (dømega/dt)ømega^-2 <<1 and if the time scale of interest is much longer than the response (L/R) time of the shell. Otherwise, the transient theory should be adopted. We begin with the study of a single representative dynamo mode, in which the steady state theory is used to evaluate the changes of the EM torque induced by the resistive shells in two variants of two RFP machines: 1) RFX and the modified RFX; 2) T2 and T2R. The transient theory has been applied to study the time evolution of the EM torque during the unlocking of a locked tearing mode in the modified RFX. In addition, the effects of non-linear mode coupling is added to the study of the interaction between the resistive walls and the multiple modes. The formation of the slinky mode and its dynamics in the presence of resistive walls are discussed.

[FP1.057] Nonlinear Resistive MHD Computations of PPCD and Cross Section Shaping in the RFP

A study of plasma dynamics during Pulsed Poloidal Current Drive (PPCD) in the reversed field pinch is conducted with non-ideal magnetohydrodynamic simulations in cylindrical and toroidal geometry using the NIMROD code [http://nimrodteam.org]. Finite pressure simulations at a Lundquist number of S\sim2000 show a marked reduction in core-resonant dynamo activity and increases in instantaneous energy confinement time commensurate with applied poloidal electric field strength (up to 66% increase in the strongest electric field case). A mode-decomposition study with varied numbers of toroidal Fourier components and an analysis of the contributions to Ohm’s law uncovers the dominant mechanisms governing the parallel current profile evolution. Computational studies at S-values greater than 2000, including a scaling study of the effects of Lundquist number on magnetic fluctuation reduction, is underway. We are also investigating how shaping the poloidal cross section affects nonlinear fluctuation amplitudes and coupling in the absence of PPCD.

[FP1.058] MHD Invariants in Turbulent Dynamo: Effect on Random-Phase Approximation and Choice of Basis Functions

We are developing a statistical treatment of an RFP MHD turbulent dynamo. This requires a closure to end the nonlinear chain of equations for the multipoint correlations. In the simplest case of homogeneous, isotropic, mirror-symmetric fluid turbulence, one focuses only on the elements of the energy spectrum because translational invariance guarantees the vanishing of , when k \neq k'. However, we find a new subtlety using a ``toy'' model of the dynamics. We numerically solve a set of equations for the evolution of the amplitudes with the form: \frac dX_i(t)dt =\sum_j,kc_ijkX_j(t)X_k(t), having two quadratic invariants: an energy and a helicity invariant. Without helicity, one finds that correlations between different amplitudes tend to vanish, as one would expect from a random-phase approximation. However, only with the set of amplitudes in which energy and helicity are simultaneously diagonal do we find this tendency when helicity is present. Extrapolating these results to the representations associated with inviscid dynamics, we expect that the Chandrasekhar basis set (i.e., the eigenfunctions of the curl operator) to be the set to which we can apply a random-phase approximation in the presence of helicity.

[FP1.059] MHD nonlinear behaviour and dynamo in numerical simulations of the reversed field pinch with external drive of toroidal magnetic flux (PPCD).

A so-called "dynamo" mechanism, due to co-operation of fluctuating velocity and magnetic fields, is usually seen as essential to describe the sustainment and transport of the Reversed Field Pinch (RFP) confinement configuration. 3D MHD nonlinear numerical simulations (SpeCyl code) are discussed in this work where an external drive is applied, similarly to the action of an experimental Pulsed Poloidal Current Drive (PPCD)[1,2]. The PPCD technique has been experimentally proven to provide substantial reduction of heat transport. In this work we show that, during the external action, the MHD (dynamo) perturbations decrease in amplitude and that the current density profiles become steeper and in particular increase in the central region, as observed in experiments [1,2]. Moreover we stress the importance of the pinch velocity term which yields the most important contribution in balancing the resistive diffusion when the turbulent dynamo becomes negligible in Ohm’s law. [1] Puiatti et al. , to be submitted to 19th IAEA 2002 conference [2] D.L.Brower et al. , Phys.Rev. Lett., 88, 185005 (2002)

[FP1.060] MHD Phenomena

[FP1.061] Local equilibrium of plasmas

What is the minimal information that must be supplied on a constant pressure surface of a toroidal plasma to completely specify a local magnetohydrodyamic equilibrium? Clearly the shape of the surface must be specified, but other information such as the separation between neighboring surfaces must also be given. We explore the conditions required for three dimensional equilibria and how local equilibrium constraints are related to global constraints. The resulting conditions provide useful checks on the accuracy of equilibrium solvers, particularly for solvers for three-dimensional equilibria, and allow local refinements in the accuracy of equilibria. Work supported by the U.S. Department of Energy under Grant No. DE-FG02-95ER54333.

[FP1.062] Simulation studies on nonlinear coupling of m/n=1/1 and 2/1 tearing modes in high-beta tokamak plasmas.

In tokamaks some major disruptions can be ascribed to nonlinear coupling of m/n=2/1 tearing mode with another tearing mode. Especially for ohmically heated high current tokamak plasmas, coupling of m/n=1/1 and 2/1 tearing modes can ergodize magnetic fields in large regions of plasmas, which would lead to significant damage to the confinement. In this work we have examined the conditions for strong coupling of m/n=1/1 and 2/1 tearing modes due to finite beta and aspect ratios, using the three-dimensional simulation code based on reduced MHD equations in toroidal and cylindrical geometries. The code uses the finite-element method(FEM) in the poloidal plane and Fourier decomposition in the toroidal direction with unstructured triangular elements and piecewise linear basis functions. The advantage of this code is that it allows us to examine nonlinear evolution of plasmas with arbitrary cross-section shapes. We shall also present results of linear stability analyses of those tearing modes using the FEM code as well as a circular-cross section cylindrical spectral code.

[FP1.063] Free boundary and Profile effects on the n=1,m=1 MHD instability

Analysis of the ideal MHD n=1,m=1, instability where n and m refer to the toroidal and poloidal mode number, is usually based on fixed boundary conditions. This effectively suppresses a source of free energy, the finite boundary perturbations. Inclusion of this effect reduces the threshold for instability. We report on details of this effect. Another aspect of the internal kink which requires detailed numerical studies is the details of the plasma profiles in the core region. Commonly accepted criteria are based only on the pressure gradients inside the q=1 surface. Here we report on the importance of the shear, q', as well as p'. An empirical criterion for stability is shown. This work was supported by DoE Contract No. DE-AC02-76-CHO-3073

[FP1.064] Nonlinear evolution of the resistive m=n=1 MHD internal mode

The macroscopic internal mode with poloidal and toroidal numbers m=n=1 plays an important role in the confinement properties of tokamak plasmas. It is the driver of the sawtooth oscillation as well as a laboratory example of fast reconnection. Meaningful fusion burn experiments are designed to operate with large plasma currents that correspond to relatively low values of the magnetic safety factor q and to a significant volume of plasma enclosed in the q=1 surface. This has a destabilizing influence on the m=n=1 mode and increases the deleterious effects of sawtooth crashes. In fusion ignition experiments, the ideal m=n=1 mode will typically be at best weakly unstable or close to ideal marginal stability due to the high values of the plasma pressure needed for fusion burning. For plasma regimes relevant to the Ignitor experiment [1], the marginal stability threshold for ideal MHD has been computed using the PEST code [1] (courtesy of J. Manickam). We focus then on those cases where resistivity controls the linear instability. We summarize the analytical results obtained in the low-\beta reduced MHD framework and present a numerical study of the onset of nonlinear effects using the 3D MHD initial value code M3D [3]. [1] B. Coppi and A.C. Coppi, Nucl. Fusion 32, 2 (1992). [2] R.C. Grimm, R.L. Dewar and J. Manickam, J. Comput. Phys. 49, 94 (1983). [3] W. Park et al., Phys. Plasmas 6, 1796 (1999).

[FP1.065] Two-Fluid Effect on Non-Linear Kink Instability in Cylindrical Tokamak with Negative Central Current

The two-fluid effect on a m/n=1/0 kink instability in a cylindrical tokamak with negative central current density is examined numerically. It is known that a m/n=1/0 resistive-kink instability removes a negative central current density, and then it flatten the current density. This flattening can explain the formation of current hole, which is a region with a very small toroidal current around a magnetic axis in strongly reversed shear tokamaks. The two-fluid effect such as electron inertia dominates this magnetohydrodynamics in high temperature tokamak plasmas. This effect induces a non-linear secondary magnetic reconnection, thereby recovering a negative central current.

[FP1.066] Integrated Predictive Simulations of Sawtooth Oscillations

The Porcelli model for sawtooth oscillations(F. Porcelli, et al.,) Plasma Phys. Control. Fusion 38, 2163 (1996), which provides a set of criteria for the triggering of sawtooth oscillations, has been implemented in the 1\frac12D BALDUR integrated predictive modeling code. In this code, sources, sinks and transport are computed self-consistently. In the Porcelli sawtooth model, sawteeth are triggered by any of three criteria for the m=1 mode: 1) kink modes when stabilization by fast ion precession fails, 2) kink modes when diamagnetic rotation stailization fails, and 3) drift tearing modes localized near the q=1 magnetic surface, when stabilization by kinetic layer and rotational effects fails. BALDUR simulations of JET and DIII-D discharges using the theory-based Multi-Mode transport model (MMM95) indicate that most of the sawteeth are triggered by the third criterion, and this criterion is found to be sensitive to the relative amount of magnetic reconnection during each sawtooth crash. The trigger is less frequent when there is more magnetic reconnection. The sawtooth period predicted by the Porcelli model in BALDUR simulations fluctuates in time, which is also observed in experiments.

\vskip4pt ^Supported by DoE contract DE-FG02-92-ER-54141

equilibrium, energy and particle transport fluxes, sources such as neutral beam heating, sinks such as impurity radiation, and the results of large scale instabilities such as sawtooth oscillations in tokamak plasmas.

[FP1.067] Extended MHD Simulations of Stellarators and Tokamaks with M3D

The M3D code has been applied to ideal, resistive, two fluid, and hybrid simulations of quasi axisymmetric stellarators and tokamaks. Ideal and resistive ballooning modes with moderate toroidal mode number (n \sim 10 ) have been studied in an NCSX design. The resistive modes occur well below the ideal \beta limit, and have growth rate scaling with resistivity \eta similar to tearing modes. When two fluid gyroviscous effects are included nonlinearly in the simulations, the resistive modes are stabilized. The two fluid simulations were done with a realistic value of the ratio of ion skin depth to major radius, c / (ømega_pi R) = 0.01. At this value high mode number ideal modes should also be stabilized, which can have a significant effect on the \beta limit. Hybrid gyrokinetic simulations with energetic particles have been done in a two - period compact stellarator geometry. A predominantly n = 1 toroidal mode was found with the frequency of a TAE mode, with growth rate linear in the energetic particle \beta. Increasing the deviation from axisymmetry (deforming a tokamak into a stellarator) has a stabilizing effect. Simulations are in progress incorporating an external vacuum magnetic field with an axisymmetric resistive wall, in order to study halo currents in ITER. The simulations include self consistent resistivity and density evolution, in which the vacuum region between the plasma and resistive wall is modeled by high resistivity and low density.

[FP1.068] Hybrid Simulations of Energetic Particle-driven MHD Mode in Toroidal Plasmas with M3D

The energetic particle-driven MHD modes are studied using hybrid model of the M3D code. The previous version of the M3D code was mainly applied to axisymmetric circular tokamak geometry. Recently we have extended the M3D code to full 3D geometry using unstructured mesh in poloidal planes and finite difference in toroidal direction. The code has also been parallelized so it can run in parallel using OpenMP or MPI. These new capabilities enable us to simulate fast ion-driven Alfven modes in advanced tokamaks, spherical tokamaks and stellarators for realistic parameters. First, fast ion-driven Alfven modes in spherical tokamaks are studied. Simulation results show an n=1 TAE mode destabilized by fast ions at low plasma beta. However, as plasma beta increases, the mode frequency reduces to a low value at about 25% of the TAE frequency. Second, fast ion-driven Alfven modes in quasi-axisymmetric stellarators are studied. For a two period stellarator, results show an unstable global TAE mode with n=1 as the dominant toroidal mode number. The effects of 3D shape is found to be stabilizing. Third, alpha-driven TAE/EPM and alpha particle effects on internal kink mode in burning plasmas are studied. Initial results indicate high-n TAE/EPM are stable in FIRE.

[FP1.069] Two-Fluid Effects on Nonlinear Steady States in Stellarators*

Nonlinear simulation with the M3D code shows that two-fluid effects are crucial to predicting steady states in the proposed NCSX ``quasi-axisymmetric'' stellarator, at realistic values of the two-fluid parameter H=1/(Ømega_ci\tau_A). In MHD at H = 0, barely unstable ideal ballooning modes grow with a resistive contribution near the low-field edge of the plasma, with toroidal mode numbers n\sim 10. Outer magnetic surfaces may be lost over a considerable radius. For H\ne 0 and p_i\ne 0, ion diamagnetic effects generate a poloidal shear flow of the plasma mass relative to the magnetic configuration and robustly suppress the modes, even at very large resistivities. If p_e\ne 0, enhanced reconnection due to electron two-fluid effects is observed in faster island formation at low order rational magnetic surfaces. At higher beta, the ion flow continues to suppress ballooning modes, but magnetic island growth is also faster and may distort the configuration sufficiently to lose the flow stabilization. The achievable steady state beta and magnetic configuration is thus limited by two-fluid effects. Similar suppression of mode growth due to ion diamagnetic effects and nonlinear enhancement of magnetic island growth by electron effects, due to the parallel gradient of the electron pressure in the Ohm's law, occurs in tokamaks. \leftline*Sponsored by the U.S. Department of Energy.

[FP1.070] NIMROD Simulations of Classical Tearing Modes

Neoclassical tearing modes have received significant interest in the experimental and theoretical communities because they significantly limit the performance of long-pulse plasma discharges. Many theoretical and experimental questions remain, one of which is the role of the \Delta' parameter, first introduced in the study of classical tearing modes. To isolate the effects of this parameter experimentally, recent experiments(M.S. Chu and R.J. LaHaye - private communication) that produce tearing modes have been performed in low pressure, Ohmic discharge on the DIII-D tokamak. Using a recent experiment with modern diagnostics and accurate equilibrium reconstruction provides an excellent opportunity for accurate comparison of theory with experiment. Preliminary NIMROD modeling of this experiment based on EFIT equilibrium show that a vacuum region is needed to model this experiment. Because NIMROD is a nonlinear, initial-value code which must allow for a moving separatrix, NIMROD models the ``vacuum'' region as a cold, highly-resistive plasma. Here, we present results from NIMROD's modeling of this experimental discharge using a three-dimensional resistivity which varies with time.

[FP1.071] Alfvén Waves in Gyrokinetic Plasmas

The ordering that \rho_i \ll L_eq enables us to use the gyrokinetic description for a magnetically-confined plasma,^1-3 by seperating the fast gyromotion of the particle from its slow gyrocenter motion and the associated polarization effects. In this paper, we present a brief comparison of the properties of the Alfvén waves based on the conventional MHD descriptions with those under the gyrokinetic approximation, and show that the unique treatment of the polarization effects in the gyrokinetic formalism makes it possible to eliminate the compressional Alfvén waves without resorting to geometric simplification for the reduced gyrokinetic-MHD equations. Thus, one may use the existing gyrokinetic particle simulation techniques^1,4-5 together with the elliptic multigrid solver^6 on the modern day massively parallel computers to study the effects of neoclassical and microturbulence transport on the global (MHD) problems based on the realistic MHD equilibria.

\$This work is supported by the US DoE SciDAC project.

\$^1W. W. Lee, Phys. Fluids 26, 556 ('83); J. Comp. Phys. 72, 243 ('87).

\$^2T. S. Hahm, W. W. Lee and A. Brizard, Phys. Fluids 31, 1940 ('88).

\$^3H. Qin, W. M. Tang and W. W. Lee, Phys. Plasmas 7, 4433 ('00).

\$^4I. Manuilskiy and W. W. Lee, Phys. Plasmas 7, 1381 ('00).

\$^5W. W. Lee, J. L. V. Lewandowski, T. S. Hahm, and Z. Lin, Phys. Plasmas 8, 4435 ('01). \$^6J. L. V. Lewandowski, manuscript in preparation.

[FP1.072] Study of Thermonuclear Alfvén Instabilities in Next Step Burning Plasma Experiments^1

A study\footnotetext[1]This work is supported by US DoE contract DE-AC02-76CH03073. is presented for the stability of \alpha -particle driven shear Alfvén Eigenmodes (AE) for the three burning plasma proposals, ITER, FIRE and IGNITOR. An analytic assessment of TAE stability is first presented and stability boundaries are determined. Then the High-n stability code HINST is used. HINST computes the non-perturbative kinetic solutions of the Alfvén eigenmodes. The stability calculations are repeated using the global code NOVAK. We show that for these tokamaks the spectrum of the least stable AE modes, TAEs, appears at medium-/high-n numbers. In HINST, TAEs are locally unstable due to the alphas' pressure gradient in ITER and FIRE. However, NOVAK calculations show that the global mode structure enhances the damping mechanisms and produces stability in all configurations considered here. A serious question remains whether the perturbation theory used in NOVAK overestimates the stability predictions, so that it is premature to conclude that the nominal operation of all three proposals are stable to AEs. In addition in ITER the beam energy is high so that the beams induce a destabilizing effect on TAEs.

[FP1.073] Excitation of Toroidicity-Induced Alfvén Eigenmodes using Electrodes in a Heliotron/Torsatron Plasma

Interaction between Alfvén waves and energetic ions is a very important issue in a fusion research. One of important issues in Alfvén eigenmodes(AE) studies is to directly measure the damping rate of AEs. For this purpose, the excitation of toroidicity-induced Alfvén eigenmodes(TAE) was attempted by externally applied magnetic perturbations in the range of 100\,\mathrmkHz in a low temperature and low density plasma of the Compact Helical System(CHS) heliotron/torsatron. The magnetic perturbations are generated by oscillatory currents that are induced along the magnetic field line by two electrodes arranged in the toroidal direction. When the frequency of the oscillatory current is swept in time, the ratio of induced magnetic fluctuations to the induced currents clearly exhibits a character of the resonance at the frequency ømega_0. The frequency ømega_0 agrees well with the TAE gap frequency. The measured damping rate \gamma_\mathrmd/ømega_0 is about 10%. This large damping rate is thought that the eigenfunction of the excited TAE extends toward the plasma edge and would interact strongly with shear Alfvén continuum there.

[FP1.074] Nonlinear Behavior of Pressure Driven Mode in a Heliotron Plasma with a Vacuum Magnetic Island

It is shown that low-m magnetic islands have significant effects on the formation of the transport barriers in recent LHD experiments. A numerical simulation code has been developed for studying nonlinear evolution of interchange mode in the presence of a vacuum magnetic island. Three field RMHD equations for stellarators are solved here. As a numerical method, two dimensional finite element method is used in the poloidal cross-section. Nonlinear pressure driven modes in a heliotron configuration have been solved in a cylindrical geometry. In this work, the (m,n)=(1,1) vacuum island is introduced with an assumed non-zero value for the boundary condition of magnetic flux, \Psi_B. The width of vacuum island may be controlled with changing the magnitude of \Psi_B. The effect of vacuum island for the growth rate of n/m=1/1 mode is destabilizing. The mechanism to change the growth rate as well as the change of eigenfunction is discussed. Also nonlinear behavior will be shown in the poster.

[FP1.075] Computation of Singular MHD Instabilities with DCON

The DCON code is in wide use for computing the ideal MHD stability of axisymmetric toroidal plasmas. It uses an adaptive numerical integrator to solve a system of ordinary differential equations for the radial dependence of complex Fourier coefficients of the normal displacement, a generalization of Newcomb's equation, from the magnetic axis to the plasma-vacuum interface. Fixed-boundary stability is determined by a toroidal generalization of Newcomb's criterion. Free-boundary stability is determined by the sign of the lowest eigenvalue of the sum of plasma and vacuum response matrices. DCON has been extended to compute the outer ideal region matching conditions for singular modes such as resistive and neoclassical tearing modes. A matching matrix is constructed from asymptotic coefficients of resonant and nonresonant solutions on either side of each singular surface and corresponding terms from any singular surface model. A dispersion relation for growth rates and eigenfunctions is obtained from the roots of the determinant of this matrix. A new, more accurate method for crossing the singular surface has avoided numerical errors affecting previous efforts to accomplish this. Benchmarks with the PEST 3 resistive stability code will be presented.

[FP1.076] AEST: Adaptive Eigenvalue Stability Code

An adaptive eigenvalue linear stability code is developed. The aim is on one hand to include the non-ideal MHD effects into the global MHD stability calculation for both low and high n modes and on the other hand to resolve the numerical difficulty involving MHD singularity on the rational surfaces at the marginal stability. Our code follows some parts of philosophy of DCON by abandoning relaxation methods based on radial finite element expansion in favor of an efficient shooting procedure with adaptive gridding. The \delta W criterion is replaced by the shooting procedure and subsequent matrix eigenvalue problem. Since the technique of expanding a general solution into a summation of the independent solutions employed, the rank of the matrices involved is just a few hundreds. This makes easier to solve the eigenvalue problem with non-ideal MHD effects, such as FLR or even full kinetic effects, as well as plasma rotation effect, taken into account. To include kinetic effects, the approach of solving for the distribution function as a local eigenvalue ømega problem as in the GS2 code will be employed in the future. Comparison of the ideal MHD version of the code with DCON, PEST, and GATO will be discussed. The non-ideal MHD version of the code will be employed to study as an application the transport barrier physics in tokamak discharges.

[FP1.077] A Computational Approach to the Hazeltine-Mahajan Two-Fluid Equations

Relativistic two-fluid equations have been recently developed by Hazeltine and Mahajan(Hazeltine, R. D., and S. M. Mahajan, Ap. J. 567) (2), 1 (2002), which extend the models of MHD and of Chew, Goldberger, and Low with a new closure procedure that is valid in regions of low collisionality and high (relativistic) velocities and temperatures and that includes parallel heat flow and pressure anisotropy as essential components. Since these equations include a richer class of phenomena than the usual MHD equations, and since both anisotropic pressure and parallel heat flow are essential components of the theory, these equations present unique numerical challenges. We present a computational plan to begin initial numerical exploration of this interesting system.

[FP1.078] MHD-like equations for electromagnetic kinetic simulations

In particle simulations of electromagnetic plasma perturbations, the perturbed fields are frequently expressed in terms of the potentials \phi and \mathbfA_\parallel. An alternative choice is \phi and \mathbfA_\perp or equivalently \mathbf\xi = ( \mathbfb \times \mathbfA )/B, with \mathbfA_\parallel = 0. The vector \mathbf\xi is related to the magnetic field line displacement and \nabla_\perp \cdot \mathbf\xi to field line compression. Plasma kinetic equations are derived in terms of these variables and Maxwell's equations are formulated so as to require the evaluation of the plasma pressure rather than the plasma current. The linearised eigenmode equations are essentially quasi-neutrality for \phi and the conventional linearised MHD fluid equations for \mathbf\xi. Kinetic and nonlinear effects are included by integrating the drift-kinetic (or gyrokinetic) equation to calculate the charge density and that part of the perturbed plasma pressure which depends on the details of the particle guiding center motion. The suitability of these equations as the basis of a particle simulation code is explored.

[FP1.079] Axisymmetric MHD Equilibria with Arbitrary Flow and Applications to NSTX

In recent years, increasing interest has been devoted to the properties of flowing plasmas. It is well known that bulk plasma rotation and sheared flow improve MHD stability (m = 1, wall mode and ballooning) and reduce turbulent energy transport. Plasma rotation can be self-generated or driven by parallel injection of neutral beams. The combination of a tight aspect ratio and fast toroidal rotation in NSTX leads to MHD equilibria that are considerably different from static ones. If the toroidal flow is of the order of the sound speed, the density and pressure are squeezed outward by the centrifugal force. Additional complications arise when pressure anisotropy and finite poloidal flows are included in the computation. For instance, when the poloidal flow is of the order of the poloidal sound speed [C_s\theta = C_sB_\theta/B], the pressure, density, and velocity profiles develop radial discontinuities at the transonic surface. Here, we present a new equilibrium code (FLOW) for the solution of MHD anisotropic equilibria with arbitrary poloidal and/or toroidal flows. When pressure anisotropy is included, the equilibrium depends on six free functions that need to be assigned as input. The code is used to determine NSTX equilibria with fast (sonic and supersonic) toroidal flow and pressure anisotropy. This work was supported by the U.S. DOE Office of Inertial Confinement Fusion under contract DE-FG02-93ER54215.

[FP1.080] Equilibrium and Stability of Flowing Two-fluid Plasmas

Improved formalisms of equilibrium and stability analyses of flowing two-fluid MHD have been developed in which the generalized vorticity of each species is introduced as a characteristic quantity. These extend the conventional single-fluid MHD. Equilibria with purely azimuthal ion flow are studied analytically and numerically. ST and CT equilibria which are relevant to the current experiments are found. These equilibria have many characteristics which can be described by the single-fluid. Criteria are found for when the single-fluid is adequate and when the two-fluid is necessary. The two-fluid is needed for equilibria with a) smaller characteristic length, b) higher beta value, and c) ion flow closer to the ion diamagnetic drift. A new relation between the perturbation of the generalized vorticity and the displacement vector for each species is found. The eigenvalue equation for two-fluid stability implies that 1) not only the magnetic field but also the generalized vorticity and 2) the electron flow as well as the ion flow may affect the stability for system with moderate size.

[FP1.081] Toroidal flow stabilization of Disruptions in High \beta Tokamaks

Disruptive high \beta tokamak plasmas can be stabilized by the addition of a sheared toroidal flow. Nonlinear simulations demonstrate that confinement in flow-free high \beta tokamaks is rapidly destroyed by growing fingers of hot plasma that jet out from the the center of the discharge to the wall. The added toroidal flow can eliminate the fingers and restore confinement. As \beta is increased further, the toroidal flow becomes less effective at maintaining a stable plasma. But a sound speed toroidal flow can increase the critical value of \beta below confinement is maintained without disruption.

[FP1.082] An eigenvalue formulation for the feedback stabilization of resistive wall modes in a toroidal geometry

A theoretical study for the feedback stabilization of resistive wall modes (RWM) is presented in a toroidal geometry. A general eigenvalue formulation describing the dispersion relation of resistive wall modes in a toroidal geometry in the presence of a set of stabilizing discrete feedback coil currents is developed. The coupling effects between the toroidal eigenmodes and a set of discrete stabilizing coil currents are included. The formalism is then applied to the proposed KSTAR plasmas and segmented FEC/RWM coil system to evaluate the maximum achievable plasma beta and power supply requirements for the feedback coil system.

[FP1.083] Two-step control of wall mode and the monodromy matrix

The Floquet stability of systems of differential equations with piecewise constant periodic coefficients is considered. In the "two-step" case the monodromy matrix is the product of two matrix exponentials. It can be evaluated either by reducing the matrix exponentials to polynomials on the eigenvalues of the corresponding matrices or, without knowledge of the eigenvalues, by using the Baker-Campbell-Hausdorff formula. The two methods are discussed and applied to several examples including the "two-step" dissipative Hill's equation introduced recently (see [1]) to stabilize the "resistive wall mode" in magnetically confined plasmas. [1] H. Tasso and G.N. Throumoulopoulos, Physics of Plasmas 9, 2662 (2002).

[FP1.084] Feedback and Control of Linear and Nonlinear Global MHD Modes in Rotating Plasmas

We present studies of feedback applied to resistive wall modes in the presence of plasma rotation. The main tool used is a Newton-Krylov nonlinear reduced resistive MHD code with completely implicit time stepping[1]. The effects of proportional and derivative gain and toroidal phase shift are investigated. In addition to studying the complete stabilization of the resistive wall mode, we present results on controlling the amplitude of nonlinear modes locked to the wall but propagating slowly; we also show results on reducing the hysteresis in the locking-unlocking bifurcation diagram.

[1] L. Chacon, D. A. Knoll and J. M. Finn, "An implicit, nonlinear reduced resistive MHD solver", J. Comp. Phys. v. 178, pp 15-36 (2002).

[FP1.085] Theoretical simulation of the feedback of the resistive wall mode

We present further results of our modeling of the feedback stabilization(M.S. Chance, et. al., \textitTheoretical modelling of the feedback stabilization of external MHD modes in toroidal geometry), Nucl. Fusion \textbf42 (2002) 295-300; M. S. Chu, et. al, \textitNormal mode approach to the modelling of the feedback stabilization of the resistive wall mode, submitted to Nucl. Fusion of the resistive wall mode in tokamaks. We have studied the effects of different sensor coil orientations, and have also made progress in exploring the advantages of placing the the feedback coils inside the resistive vacuum vessel. The formulations and results will be presented.

[FP1.086] Dynamic Responses of Tokamak Plasmas Induced by Externally Applied Rotating Helical Magnetic Field in Dynamic Ergodic Divertor

Dynamic Ergodic Divertor (DED) is an advanced concept for the control of the tokamak edge plasmas. In contrast to conventional Ergodic Divertor, an externally applied helical magnetic field rotates in the helical direction. This rotating helical field (RHF) may decrease the heat and particle flux onto the local target in the conventional divertor. In the DED experiment, it is also expected that RHF induces the edge plasma rotation due to the shielding current around the resonance surface. Penetration processes into tokamak plasmas and dynamic behavior of RHF have been investigated on HYBTOK-II tokamak. We have confirmed the attenuation of the radial component of RHF from the magnetic probe measurement. When the relative rotation velocity between the plasma and RHF is small around resonance surface, however, the radial component of RHF is amplified by the effect of re-distribution of the plasma current by island formation. We will analyze the penetration process of RHF by linear MHD theory.

[FP1.087] Two-Dimensional MHD Simulations of Tokamak Plasmas with Poloidal Flow

A two- dimensional MHD code has been developed to simulate the temporal evolution of Tokamak plasmas with an imposed poloidal flow. The code is fully compressible and can resolve the shock structures arising when the poloidal velocity is of the order of the poloidal sound speed (V_\theta\sim C_s B_\theta/B) near the plasma edge, where the plasma is cold and the sound speed is low. The poloidal flow is assigned as an initial condition with a velocity profile ranging from subsonic to supersonic near the edge. It is found that a continuous band of shocks is formed near the edge. Such shocks travel poloidally, leaving behind a pedestal structure similar to the one predicted in Ref. 1 [R. Betti and J. P. Freidberg, Phys. Plasmas 7, 2439 (2000)]. Here, the pedestal is defined as a sharp discontinuity in the pressure, temperature, and density profiles. The pedestal height is modulated in the poloidal angle; it is maximum on the outboard side (\theta = 0) and minimum on the inboard (\theta = \pm\pi). Furthermore, both poloidal and toroidal flows develop a shear layer at the location of the pedestal. The large velocity shear (both poloidal and toroidal) occurring in the pedestal region is likely to suppress turbulent eddies and reduce anomalous transport. This work was supported by the U.S. Department of Energy Office of Inertial Confinement Fusion under Cooperative Agreement No. DE-FC03-92SF19460.

[FP1.088] Magnetic Island Induced Bootstrap Current and Its Implications

It is shown that a magnetic island in a one-dimensional equilibrium plasma, such as a cylindrical plasma or a slab plasma can induce a pressure-gradient-driven current parallel to the magnetic field similar to the bootstrap current in toroidal plasmas. This current flows on both sides of the island in opposite directions. It modifies the current density profile that drives the island growth in the vicinity of the island and may provide a mechanism for island saturation in such plasmas. The magnetic island induced plasma viscosity damps the helical component of the plasma flow velocity. This alters the polarization drift. The polarization drift also has opposite signs on both sides of the island. The plasma electrical conductivity is reduced due to the existence of trapped particles when the island width is finite. The implications on tokamak neoclassical island will be discussed.

[FP1.089] Nonlinear evolution of shear-localized ideal interchange instability

Suydam/Mercier criterion demarcates the instability boundary of a toroidally confined plasma for a radially-localized pressure-driven ideal MHD interchange instability. Many present day tokamaks operating in high performance regimes (i.e., high \beta) are very close to the ideal stability boundary. In this work, we address the nonlinear evolution properties of Suydam modes as the stability boundary is violated. Linear stability analysis of the pressure-driven ideal interchange instability indicates a very weakly growing mode since the growth rate is negligibly small well above the marginal stability condition (i.e., Suydam criterion)(Violating Suydam criterion produces feeble instabilities, Sangeeta Gupta, J. D. Callen and C. C. Hegna, report UW-CPTC 01-7 (to be published in Phys. Plasmas.)). The nonlinear behavior of these weakly growing modes is still an unanswered question. For studying the nonlinear fate of these modes, the nonlinear vorticity equation coupled with the ideal Ohm's law and pressure evolution equation are solved numerically using a Galerkin representation. Here, only the equilibrium, fundamental growing mode and its second harmonics are evolved in both space and time. Numerical results indicate that at later times a large current density appears around the mode-rational surface. The role of various nonlinearities, especially the nonlinear term in the Ohm's law, in the generation of this large current density will be discussed.

[FP1.090] Hybrid MHD Ballooning Instabilities?

Low mode number (n < 3) instabilities in toroidal plasmas are governed by ideal, resistive or neoclassical MHD descriptions in which both the electrons and ions have fluid-like responses. In contrast, very high mode number modes (n > 30) have near-adiabatic electron responses, are governed by kinetic descriptions, and usually are of the drift wave type. In this work we explore a ``hybrid MHD" description of medium mode number modes (3 < n < 30) in which the ions are always fluid-like but the electrons are kinetic on short parallel scale lengths (within one connection length \sim R_0 q) but fluid-like on long scale lengths (>> R_0 q). The key to such a hybrid description is a Chapman-Enskog-type, ``neoclassical-like" kinetic derivation of the collisional (ømega \leq \nu_e) electron response for such mode numbers. The stress and heat flow moments of this kinetic electron response yield the parallel viscous force and heat flux moments that are needed to close the fluid moment equations. These closure moments include electron Landau and flow damping effects. Hybrid MHD ballooning instabilities are then sought using a combination of the pressure balance, parallel Ohm's law and quasineutrality (or current-continuity, \nabla \cdot J = 0) equations.

[FP1.091] On the Dynamics of Taylor Relaxation

We investigate the dynamics of Taylor relaxation with the aim of elucidating general/universal characteristics. The argument of Boozer is extended to derive a model-independent 1D nonlinear pde for the transport of helicity density during Taylor relaxation in an inhomogeneous system (i.e. an R.F.P.). Imposing joint reflection symmetry further constrains the form of the helicity flux, and yields, in the large scale limit a simple, 1D equation for the time evolution of deviations of the current profile from the Taylor state. This equation extends the time-honored concept of hyper-resistive diffusion, and is perhaps the minimal model of Taylor relaxation dynamics. We have examined the implications of the model, in both a `coherent' and `stochastic' framework. In the coherent case, it is easily shown that all localized current perturbations decay time asymptotically. However, stationary current dipole pairs and traveling wave solutions exist. In the stochastic context, we derive a scaling relation between the effective magnetic Reynolds number and the noise intensity. This scaling is being compared with astrophysical data. The relation of traveling wave solutions and avalanches to relaxation transients will be discussed.

[FP1.092] MHD dynamics of relaxed plasmas sustained by Oscillating Field Current Drive

The MHD dynamics of ac magnetic helicity injection is investigated using 3-D computation at Lundquist numbers up to 500, 000 in the Reversed Field Pinch (RFP). To date most of the MHD computations have been based on the dc helicity injection. We model Oscillating Field Current Drive (OFCD, also called F-theta pumping), a form of ac helicity injection, in which toroidal and poloidal oscillating voltages are applied at the plasma surface to inject magnetic helicity and sustain steady-state plasma current. A 3-D nonlinear, resistive MHD code (the DEBS code) is used to examine the penetration of driving oscillating fields, the response of the both oscillating mean profiles and the helical tearing instabilities during a cycle, and the cycle averaged response. We study both current sustainment by oscillating fields when mean toroidal voltage is zero (helicity replacement) and partial current sustainment in the presence of ohmic drive (helicity addition). In the absence of tearing fluctuations (1-D response) a steady-state axisymmetric edge current is driven by the dynamo effect of the axisymmetric oscillating fields. It is shown that in the 3-D computation, the total toroidal plasma current can be fully sustained by OFCD helicity replacement. With the full 3-D response, resisitive MHD fluctuations (turbulence) relax the driven edge current toward the center (by the tearing mode dynamo), resulting in a steady-state current over the entire plasma cross-section. Partial current sustainment can be obtained by OFCD helicity addition. The dependency of the F-theta oscillations on frequency, as well as the oscillating amplitudes dependency on Lundquist number, S, are shown. Although the computations show that oscillating fields can drive significant amount of current, because of the excitation of the edge modes, fluctuations are enhanced during part of the cycle. We show that the resonant edge modes are excited linearly due to the large oscillations in the mean profiles. Using the field line trajectories, the magnetic field line stochasticity is also investigated for the current sustainment by oscillating fields.

[FP1.093] Magnetic Structures within Flowing Plasmas

Based on the symmetry between plasma flow and magnetic field, a new physically possible assumption about internal plasma current distribution is used to show that magnetic structures with closed flux surfaces can exist in flowing plasmas. The introduced assumption and the resulted new magnetic structures give new meanings and perspective to an ideal MHD framework initiated by Chandrasekhar in the 1950s. It is argued that the new assumption is not a natural result of the variational principle. The proposed magnetic structure family includes well-known prototypes, such as Taylor-state spheromaks, field-reversed configurations (FRCs), and an infinite number of other new members. New examples are discussed in detail. To obtain these magnetic structures, the role of plasma flow is critical. Because of the non-vanishing plasma flow, the proposed magnetic-structure-inbedded plasma states are different from conventional ones. For example, the new spheromak solutions can have non-vanishing electromagnetic force, finite plasma pressure gradient, and plasma flow, while its conventional counterpart is a force-free state with zero pressure gradient and no plasma flow. Based on these new structures, new fusion schemes that rely on the plasma flow and magnetic field self-organization might be possible. New interpretation of fusion experimental data is also possible using these proposed magnetic structures. In addition, the proposed magnetic structures might find their applications in explanation of cosmic magnetic objects.

[FP1.094] Investigation of the initial inductive startup in a tokamak plasma

In the tokamak startup phase, a large temporal magnetic flux change is required for having sufficient loop voltage inside the vacuum vessel. The large flux change causes large eddy currents flowing in various conducting structure materials such as vacuum vessel, passive stabilizer, toroidal conducting loops for compensating the connection of adjacent toroidal field coils, etc. The vacuum magnetic field structure may be significantly influenced by the eddy currents. For efficient inductive startup, it is important to control the field geometry especially at the initial startup phase in a tokamak. With an exemplary tokamak geometry, we studied the generation of field null inside the vacuum vessel including several driving currents and induced eddy currents in structural materials. Under such circumstances, the evolution of plasma parameters such as density, temperature, and plasma current is investigated through numerical simulation.

[FP1.095] Free-boundary 1 amp; 1/2D Simulation of Disruptions in ITER-FEAT.

During the current quench (post-thermal) of a disrupting plasma, large currents induced in the plasma halo, closing via poloidal flow through conducting structures, can produce large, damaging forces on the vacuum vessel and plasma-facing components. Predictive models are necessary to evaluate these forces. An analytic model of the current quench, consisting of a pair of ODE's in time for the core and halo currents, has been shown to agree well with data from DIII-D; measurements of T_e and Z_e\mkern-3mu f\mkern-3mu f in an experiment with massive injection of He gas confirm that post-thermal-quench resistivities are classical(D.A.~Humphreys and D.G.~Whyte, Phys.~Plasmas 7), 4057 (2000).. Here we develop a 1 amp; 1/2 D implementation of this model for use in our transport code Corsica(http://wormhole.ucllnl.org/caltrans/); the upgrade of the code to include modelling of current transport on the open field-lines is in progress. We will apply the code to ITER-FEAT, and various disruption-mitigation scenarios using high-pressure noble-gas injection will be explored. Comparisons with the 0-D model will be presented.

[FP1.096] Dynamics of toroidally coupled magnetic perturbations and seed magnetic island formation in tokamak plasmas

A model is developed in order to describe the dynamics of the magnetic island 'seeding’ process required for the initiation of a neoclassical tearing mode in tokamak plasmas. The formation of a magnetic island due to a toroidally coupled magnetic perturbation is modeled as an initial value problem. If differential rotation exists between the two magnetic surfaces of interest, the dynamics of the flow profile evolution also needs to be accounted for. By extending the model beyond resistive MHD, the role of two-fluid and neoclassical physics on magnetic island formation can be addressed. In the presence of a fast growing external magnetic perturbation and differential rotation, the effect of the neoclassical polarization mechanism on establishing a nonlinear island width threshold is altered from cases appropriate for long resistive island evolution timescales. In order to obtain a complete solution to this problem in general, a neoclassical viscous force that accounts for time dependent processes (time scales faster than the ion collision time) needs to be developed. Implications for neoclassical tearing mode dynamics in high temperature plasmas will be discussed.

[FP1.097] Dusty and Non-Neutral Plasmas

[FP1.098] ROLE AND PROPERTIES OF FORMED SOLITARY ELECTRIC FIELD IN DUSTY PLASMA FLOW WITH FULLERENES IN SOURCE OF NANOTUBES

The plasma flow with positive ions and electrons propagates relative to heavy negative ions of fullerenes in source of nanotubes. The flow excites the solitary electric potential dip of large amplitude with potential jump near the dip. The properties and evolution of this excited perturbation are considered. The evolution equation is derived for the case of any amplitude. It is shown that this perturbation of large amplitude lead to acceleration of heavy negative ions of fullerenes from region of their generation. One can use this electric potential dip with potential jump near it for energy control of fullerenes and rate optimization of nanotube generation.

1.W. Oohara, R. Hatakeyama, S. Ishiguro, and N. Sato: Proc. 1998 Int. Cong. Plasma Physics, Praha, 22C (1998) 2419.

[FP1.099] 3D Structural and dynamical measurement of a complex plasma crystalization

Complex plasmas are often used as a model system for solid-liquid-gas phase transitions. Usually, these transitions has been studied by melting the plasma crystal by lowering the neutral gas pressure. The lower gas pressure enables instabilities to overcome frictional damping, thus heating the particles and melting the crystal. However, lowering the pressure also significantly alters the plasma environment which in turn effects the particle interaction potential.

Here, the initial stages of the crystallization of a 3D complex plasma are analyzed in a non-changing plasma environment (neutral gas pressure and plasma power constant). A 3D imaging method bas been developed in order to analyze the dynamics of crystallization in all three degrees of freedom simultaneously. The technique provides both the 3D particle position and 3D velocities in the viewed volume. The first empirical results for the time evolution of the particle temperature and microdynamics of the system will be presented.

[FP1.100] A Nonlinear Theory of Void Formation in Colloidal Plasmas

Recently, a number of colloidal plasma experiments, in laboratory as well as under microgravity conditions, have shown the spontaneous development of voids. A void is typically a small and stable centimeter-size region (within the plasma) that is completely free of dust particles and characterized by sharp boundaries. In the laboratory, the void is seen to develop from a uniform dust cloud as a consequence of an instability when the dust particle has grown to a sufficient size. We propose a nonlinear time-dependent model for void formation in colloidal plasmas. For experimentally relevant initial conditions, the model describes the nonlinear evolution of a zero-frequency linear instability that grows rapidly in the nonlinear regime and subsequently saturates to form a void. A number of features of the model are shown to be consistent with experimental observations under laboratory and microgravity conditions.

[FP1.101] Dust grain charging and levitation in a weakly collisional DC sheath

Monodisperse dust grains have been levitated above a biased conducting surface within a DC sheath in argon plasma. The observed levitation heights and their dependence upon surface bias voltage are near to values calculated from a model. The ion charging current is calculated assuming that the ions have a single energy determined by the sheath potential profile and the electron charging current is found assuming Maxwellian electrons. The grain charge is calculated from the grain potential relative to the potential in the surrounding sheath. The potential profiles are found from both the Bohm collisionless model and Riemann's collisional model. The potential profile is different for graphite and steel surfaces suggesting that secondary electrons are important.

[FP1.102] Nonlinear compressional waves in a 2D dusty plasma crystal: Theory

A plasma crystal is a strongly-coupled dusty plasma, with charged micron-size polymer spheres that arrange themselves in a pattern, like a crystalline lattice. Here, we model experiments with a 2D triangular lattice with hexagonal symmetry. The particles interact through a Yukawa repulsion. Linear compressional waves in this lattice obey a dispersion relation(X. Wang et al., Phys. Rev. Lett. 86, 2569 (2001).) which is dispersionless for long wavelengths; therefore, these waves can exhibit nonlinear effects such as three-wave mixing and harmonic generation. In experiments, waves are also damped by gas drag. Modeling the lattice as a linear chain in the continuum limit, the particle velocity v obeys a KdV-like equation, \frac\partial v\partial t + \nu_d v = -v_p \frac\partial v\partial x -\fracA2v \frac\partial v\partial x, where A depends on \kappa which is the ratio of interparticle distance and the Debye length, while v_p and \nu_d are the wave phase velocity and the gas damping rate, respectively. Beat waves and harmonics are produced, with a threshold laser power determined by the damping level. It is found that lattices with larger inter-particle separations exhibit a stronger nonlinear interaction. Work supported by NASA and DOE.

[FP1.103] Nonlinear compressional waves in a 2D dusty plasma crystal: Experiment

Experiments were performed to test our theoretical predictions of three-wave mixing and harmonic generation of compressional waves in a 2D plasma crystal. A 2D triangular lattice with hexagonal symmetry was formed by levitating a monolayer of particles in the sheath of a gas-discharge plasma. The sheaths curvature provides a bowl-shaped external confining potential. The particles were imaged using a video camera, to record their motion. We excited waves in the lattice using the radiation pressure of a sheet of argon laser light. Waves propagated away from the narrow excitation region. The amplitude of the wave excitation is adjustable by varying the laser power. The wave damping is determined by the gas pressure. We adjust the particle spacing by varying the number of particles and the discharge conditions. Particle motion was observed identifying particles in each image and computing their velocities. We excited waves at two frequencies, using two laser sheets. The power spectrum of the waves is computed to identify beat waves at the sum and difference frequencies, as well as harmonics.

[FP1.104] Nonlinear compressional waves in a 2D dusty plasma crystal: Simulation

Molecular dynamics simulations were performed to study three-wave mixing and harmonic generation of compressional waves in a 2D plasma crystal. These simulations were used to verify our analytic theory based on a KdV-like equation, and to predict optimal experimental parameters, in preparation for laboratory experiments using a dusty plasma. Using 5000 particles, which interact with a Yukawa potential and are trapped by a parabolic external confining potential similar to the bowl-shaped electrode sheath in the experiment, we integrate the particles equation of motion for 60 seconds. Applying a local external force, modulated in time to model the radiation pressure force applied by a laser in the experiment, sinusoidal compressional waves are excited. Using two excitation regions at different frequencies, mixing occurs at the sum and difference frequencies as well as harmonic generation. We test the analytic theory’s predictions of a threshold and of a spatial variation of beat wave amplitude with distance from the excitation region.

[FP1.105] Phase Transition in Dusty Plasmas: A Microphysical Description ^*

We report on analytical and simulation studies of microphysical processes that trigger phase transitions in a dusty plasma subject to ion streaming. For pressures below the critical pressure Pc for condensation, the grains acquire a large random kinetic energy and form a weakly coupled fluid. If P is increased to greater than P_c, the grains lose their kinetic energy and reach a strongly coupled crystalline state. The dust heating in the fluid phase is due to an ion-dust two-stream instability, which is stabilized at P > P_c by the combined effect of ion-neutral and dust-neutral collisions. If one starts from the crystalline state and decreases the pressure to below the critical pressure Pm for melting, transverse phonons are destabilized by ion streaming, which destroys the short range ordering of the dust grains and triggers melting. It is found that P_m < P_c. For P_m < P < P_c mixed phase states can exist.

*Supported by Office of Naval Research and NASA

[FP1.106] Measurement of charged aerosol particles in the mesosphere by a rocket-borne probe

A magnetically shielded, charge collecting rocket probe was used on two flights of the MIDAS (MIddle Atmosphere Dynamics and Structure) SOLSTICE (Studies of Layered STructures and ICE) 2001 rocket campaign from Andoya, Norway. The probe was a graphite collection surface with a permanent magnet underneath to deflect electrons. The first launched June 17, 2001 was into a strong, multiply layered PMSE detected by the ALWIN radar. The probe measured negative particles with a peak charge number density of -1500 charges per cc. The second (June 24), into another strong, multiply layered PMSE, saw a band of positive particles centered in the lowest radar echo maximum, and a negative particle layer accompanied by a positive ion excess. The charge number densities for the PMSE particles were several thousand charges per cc. Unexpectedly, 2 km beneath the PMSE, the probe also found a very pronounced negative layer which was probably a noctilucent cloud. Computer simulations of incoming, singly negatively charged ice aerosols were performed using a rarefied flow field representative of the MIDAS payload at zero angle of attack. Ice aerosols < 1 nm in radius were diverted by the leading shock front, indicating the smallest detectable ice aerosol by this probe

[FP1.107] Interaction of Small Bodies with the Atmosphere

We consider some aspects of meteor physics. We focus on several processes relevant to the interaction of meteoroids with the atmosphere, that involve concepts typical of dusty plasmas. The objective of our research is to simulate the meteor flight directly using the kinetic PIC simulation code DEMOCRITUS [1] and a new high accuracy PIC code designed to simulate meteoroids in spherical geometry. The simulations will investigate the physics determined by thermionic emissions [2] and by the presence of potential wakes in the meteor trail. These two aspects will be considered particularly with respect to their relevance to observations, such as the solution of still outstanding problems in the interpretation of radar echoes from the head and trail of meteors.

[1] G. Lapenta, Phys. Plasmas, 6, 1442 (1999).

[2] G. Sorasio, D.A. Mendis, M. Rosenberg, Planetary Space Sci. , 49, 1257 (2001).

[FP1.108] Controlled interactions of microparticle clouds in a dc glow discharge dusty plasma

Recent investigations of dusty plasmas have shown that it may be possible to use the trajectories of charged microparticles to probe the electrical properties of a microparticle cloud and the surrounding plasma environment. In past experiments, self-generated dust particle ‘streams’ [E. Thomas, Jr., Phys. Plasmas, 8, 329 (2001)] have been used as the source of probe particles. For the experiments described here, two microparticle clouds are suspended in a plasma. The clouds are separated by an array of small, separately biased electrodes. Bias voltages applied to the electrodes are used to control the flow of particles from one cloud to the other. The particle image velocimetry (PIV) technique is used to characterize the transport of microparticles from one cloud to the other. Results are presented on the interaction of the clouds as a function of the bias voltage on different array elements. The applicability of this technique as a diagnostic approach for dusty plasmas will also be discussed.

[FP1.109] Probe induced voids in a dusty plasma

Voids are regions within a dusty plasma in which there is a complete absence of dust particles. The void boundary is characterized by a very sharp gradient in the dust density. Voids are a common feature of dusty plasmas formed under microgravity conditions, but they can also be produced by electrodes (probes) inserted into the dusty plasma. We present results of a laboratory experiment designed to study the formation of voids that are produced by biased probes located within a dusty plasma. The experiment was performed using the Auburn Dusty Plasma Experiment (DPX) which generates a dusty plasma using a dc glow discharge in argon. The voids are formed by inserting a small wire electrode into the cloud. Measurements of the size of the void were made for various bias voltages on the wire. The results will be compared with a model in which a stable void is maintained by balancing the outward electric force on the negatively charged dust particles with the inward ion drag force due to ions drawn by the probe.

[FP1.110] Expansion Rate Scaling and Diagnostic Development on the Electron Diffusion Gauge Experiment.

The expansion rate dependence on pressure of the Electron Diffusion Gauge (EDG) pure electron plasma resulting from collisions with background neutral gas atoms is analyzed. Expansion rate data is obtained for smaller initial plasmas generated with a smaller filament (outer diameter 1/4 of the wall diameter) installed in the EDG device, and the data is compared with previous results for larger-filament plasmas. The measured expansion rate in the higher pressure regime is found to be in agreement with the classical estimate \[ \fracddt\langle r^2 \rangle = \frac2 N_L e^2 \nu_enm ømega_c^2 \left(1+\frac2TN_L e^2\right). \]

Progress on the fabrication and installation of a standard on-axis parallel temperature diagnostic and a phosphor-screen-based density imaging diagnostic is also presented. The imaging diagnostic will include a grounded grid between the trap and the 3kV-biased phosphor screen to minimize azimuthal shearing of the plasma when large-amplitude m=1 diocotron modes are present. These diagnostics will help clarify the behavior of the plasma during the background-gas-induced expansion.

[FP1.111] Cryogenic Positron Plasma in a 5 Tesla Field

We present the first experimental results from a new Penning-Malmberg trap using a 5\,Tesla magnetic field and a cryogenically cooled electrode structure (T\sim5\,K) with the goal of producing cold (\Delta\epsilon\sim1\,meV), high-density positron plasmas. Positron bunches will be delivered in short bunches from a separate accumulation trap that is fed by a radioactive source and a solid neon moderator [1]. Using a 100\,mCi ^22Na positron source, filling rates of 10^10 e^+/h and plasma densities of 10^9\,cm^\mbox-3 are expected. The positrons will thermalize to less than 10\,K by cyclotron radiation. Radial compression is achieved by applying a rotating electric field [2]. This trap is designed to be a nearly ideal reservoir of positron plasmas, with very long confinement and annihilation times. The trap design, operation, its potential uses and first experimental results with electron plasmas will be discussed. [1] C.M. Surko, \textitet al., Non-Neutral Plasma Physics III, J.J.\,Bollinger, \textitet al., eds., American Institute of Physics (1999), pp. 3--12.

[2] R.G. Greaves and C.M. Surko, Phys. Plasmas, \textbf8, 1879 (2001).

[FP1.112] Design of a High-Capacity Penning-Malmberg Trap for Positrons.

There are a number of motivations to accumulate large numbers of positrons [1]. We describe a multicell Penning-Malmberg trap capable of confining \mboxN = 10^15 positrons, based on current knowledge of cross-field transport of electrons in such traps. A novel multicell design helps to minimize space charge limitations. Heating associated with plasma expansion is balanced by cyclotron cooling in a 10\,T field, and long-term confinement is achieved using a "rotating wall" electric field [2]. Confinement scaling and avoiding ionization on background gas place important constraints on the plasma parameters: \mboxn \sim 10^11\,\mboxcm^\mbox-3;\; \mboxT = 6\,\mboxeV;\; \mboxN/cell \sim 10^10;\; \mboxL_\mboxp = 1\mboxcm;\; \mboxr_\mboxp = 1.5\,\mboxmm. A trap for 10^15 positrons using 10\,kV potential wells requires 10^5 cells and occupies a volume 1\,m in length by 0.5\,m in diameter. Applications, such as portable positron sources will be discussed, as well as the research program required to implement this design.

[1] R. G. Greaves and C. M. Surko, Phys. Plasmas, \textbf4, 1528 (1997).

[2] E. Hollmann, \textitet al., Phys. Plasmas, \textbf7, 2776 (2000).

[FP1.113] Two-Stream Interactions in Guiding-Center Plasmas for Antihydrogen Recombination Schemes

Two-stream instabilities are studied analytically in the guiding center kinetic regime, which is chosen in order that the results may be applied to the mixing of antiprotons and positrons preceding antihydrogen recombination. The guiding center kinetic description is valid for a range of parameters which includes cases in which magnetic fields are 3-5\,\rmT; temperatures are 4-10\,\rmK; positron densities are 10^7 - 10^8\,\rmcm^-3; antiproton densities are 10^4 - 2\!\times\!10^7\, \rmcm^-3; the positron column radius is 0.05 - 0.3 wall radii; and the antiproton column radius is 0.05 - 0.1 wall radii. The species occupy long, cylindrical columns coaxial with an outer conducting cylinder. A constant, axial, externally generated magnetic field permeates the system. Linear stability of the counter-streaming Maxwellian plasmas is found as a function of the species' temperature ratio, density ratio and mean relative velocity. Preliminary results for the effect of positron-positron collisions on stability behavior are discussed.

[FP1.114] Progress on new temperature measurements and excitation of shear modes in Penning trap ion crystals

We describe current experimental efforts to study shear modes and measure heating rates in ion crystals confined in a Penning trap. Up to 10^6 ^9Be^+ ions are laser-cooled to less than 10 mK where they form a crystalline state. Previously, the ion temperature was measured from the Doppler broadening of an optical transition. Measurements below 10 mK were inaccurate due to the 19 MHz natural linewidth of the transition. We will summarize our progress on using a stimulated Raman transition between two ground-state hyperfine levels to more accurately measure the ion temperature. The width of this transition does not depend on the linewidth of the excited state. Measurements of the ion heating are motivated by the possibility of using the Penning trap for making long-lived entangled and spin-squeezed states of ions. We will also summarize our attempts to excite and detect shear modes. Shear modes exist only in a crystalline (as opposed to a liquid) plasma and provide a sensitive probe of the ion correlations. As the lowest frequency modes of the system, they could be an important factor in determining the usefulness of the Penning trap for quantum information experiments.

[FP1.115] Isolating Diffusive Effects in Asymmetry-Induced Transport

In our studies of asymmetry-induced transport in a modified Malmberg-Penning trap, a typical data set consists of the radial flux \Gamma vs radius r for many values of the asymmetry frequency ømega. For a given asymmetry frequency, the flux has a complicated radial dependence, presumably because contributions from diffusion, mobility, and resonant particle effects all depend on r in different ways. To build an empirical model of this transport, we would like to study each of these contributions in isolation. According to the theory(D.L. Eggleston and T.M. O'Neil, Phys. Plasmas 6, 2699 (1999).) both the mobility and the resonant particle factor contain the quantity ømega - lømega_R, where ømega_R(r) is the E\times B rotation frequency and l is the azimuthal mode number of the asymmetry. In an attempt to isolate the diffusive transport, we have thus selected from our \Gamma vs r vs ømega data set those points where ømega matches lømega_R. The resulting plots of \Gamma vs density gradient \nabla n show a simple relationship which is largely independent of center wire bias (as expected), but which exhibits significant deviations from linearity. This may indicate that other effects (e.g. radial temperature gradients) play an important role in the transport.

[FP1.116] Numerical Simulation of Ultracold Plasmas.

In recent experiments ultracold neutral plasmas were produced by photoionizing small clouds of laser cooled atoms. This paper presents the results of molecular dynamic simulations for the early time evolution of such plasmas. Contrary to earlier speculation, no evidence of strong electron-electron correlations is observed in the simulations even if the initial value of the coupling parameter (\Gamma_e = e^2 / akT_e) is much larger than unity. As electron-electron correlations begin to develop, the correlation energy is released to heat the electrons, raising the electron temperature to the point where \Gamma_e \sim 1 and limiting further development of correlation. Further heating of the electrons occurs as a by-product of three-body recombination. When a model of laser cooling is added to the simulation, the formation of strong ion-ion correlation is observed. The rate of three-body recombination is observed to be in reasonable agreement with the traditional formula, R = 3.9 \! \times \! 10^-9\:sec^-1 [n \; (cm^-3 )]^2 \; [ T_e (^\circ K)]^-9/2, but care must be taken to use the correct temporally evolving temperature, T_e. The simulations are challenging because it is necessary to follow three-body recombination into weakly bound (high n quasi-classical) Rydberg states, and the time scale for such states is short compared to that for the plasma dynamics.

[FP1.117] Shear-Limited Test Particle Transport in Two-Dimensional Plasmas.

Measurements of test-particle transport in pure ion plasmas show 2D enhancement over the 3D rates, limited by the shear(C.F. Driscoll et al.), Phys. Plasmas 9, 1905-1914 (2002). in the E \times B plasma rotation ømega_E. For finite plasma length L_p, thermal particles may bounce axially many times before rotational shear separates them in \theta. This number of bounces N_b \equiv ( \barv / 2 L_p) / (r \partial ømega_E / \partial r ) characterizes the approach to the 2D bounce-average regime. For N_b < 1, test particle diffusion is due to long range E \times B drift collisions with impact parameters in the range r_c < \rho < \lambda_D. Over the range 1 < N_b < 100, experiments measure test particle diffusion increasing as N_b. For exceedingly small shear N_b > 1000, we observe transport rates consistent with the Taylor-McNamara estimate for shear-free thermal plasmas. Experimental data suggest the existence of convective transport superimposed on diffusion, consistent with the theory idea of transport due to large thermally excited vortices.

[FP1.118] Shear-Reduction of Collisional Transport in a 2D Point Vortex Gas/Plasma.

The two dimensional point vortex gas is a simple but useful paradigm for more complex fluid and plasma flows.(D. Dubin and D. Jin, Phys. Lett. A 284), 112 (2001). This poster presents the theory of the collisional diffusion and viscosity coefficients for a point vortex gas, in an applied shear flow. We show that the transport coefficients are reduced in the presence of shear, just as for the shear reduction of transport observed in fusion plasmas. Here however, fluctuations are collisional rather than turbulent, allowing a rigorous calculation of the transport. When there are several species of point vortices, we find that Onsager relations require that the diffusive flux conserves the total vorticity density \rho (r) (proportional to charge density in the plasma analogue). Surprisingly, the diffusive flux concentrates vortices with large positive (or negative) circulations at maxima (or minima) of \rho (r). On a slower timescale, the momentum flux due to viscosity drives the system to a global thermal equilibrium state.

[FP1.119] A Numerical Study of the Influence of Dynamics on 2D Turbulent Relaxation.

Over the years, several conflicting theories have been put forward to predict the final state of freely relaxing inviscid 2D turbulence in fluids and magnetized plasmas. Theories based on minimizing enstrophy, maximizing Boltzmann entropy, maximizing fluid entropy, or maximizing the entropy of a system with persistent vortices (yielding vortex crystal states) all lead to different predictions for given values of the robust invariants (the energy, angular momentum and circulation). In order to explore the utility of these theories, we ran many numerical vortex-in-cell (VIC) simulations, starting with unstable initial conditions with the same values of the robust invariants, but with different vorticity profiles. Our results show that the robust invariants by themselves are generally poor predictors of the final relaxed state. We observe sensitive dependence of the final state on initial conditions, thereby emphasizing the importance of dynamics on relaxation. Our findings suggest three general classifications of final states: single central vortex, vortex crystal, and minimum enstrophy. In general, dynamics dominated by vorticity excesses (clumps) lead to vortex crystals or a single central vortex; whereas dynamics dominated by vorticity deficits (holes) lead to minimum enstrophy-like final states.

[FP1.120] Observation of Diocotron Wave Echoes in a Pure Electron Plasma.

We have observed diocotron wave echoes in magnetized electron columns, demonstrating the reversible nature of spatial Landau damping. Here, two diocotron waves are externally excited separated in time, and a third diocotron wave (the echo) appears after the two applied waves have inviscidly damped away. Experimental images(J.H. Yu and C.F. Driscoll, IEEE Trans. Plas. Sci. 30), 24-25 (2002). show the damping as the spiral wind-up of the density perturbation, and show the unwinding which results in the echo. The diocotron wave phase mixing (and unmixing) can be observed directly because the phase space ( \theta , p_\theta ) is equivalent to the configuration space ( \theta , r^2 ). Experiments agree with theory on the angular mode number and appearance time of the echo. The echo is surprisingly robust, occurring even though end effects cause a v_z-dependence in the E \times B rotation of each electron. Only at late times does collisional velocity scattering destroy the dynamical reversibility and diminish the echo.

[FP1.121] A Theory of Trapped-Particle Asymmetry Mode Damping.

We have identified the damping mechanism of a recently discovered trapped-particle mode(A. A. Kabantsev et. al.), Phys. Rev. Lett. 87, 5002 (2001). as velocity scattering in a boundary layer near the separatrix. This mechanism is similar to that invoked earlier to describe the damping of the dissipative trapped-ion mode.( M. N. Rosenbluth et. al.), Nucl. Fusion 12, 3 (1972). The new mode is excited on a nonneutral plasma column in which classes of trapped and passing particles have been created by the application of an electrostatic potential barrier. Trapped particles on either side of the barrier execute E \times B drift oscillations that are 180^\circ out of phase, while passing particles move along the field lines and Debye shield the perturbed trapped-particle charge density. A Fokker-Planck analysis focusing on a thin boundary layer near the separatrix yields a damping rate that scales as \sqrt\nu and is in good agreement with measurements. In the experiments, the damping can be enhanced by applying a weak potential that oscillates in resonance with the bounce motion of the marginally trapped particles, thus confirming that the damping is associated with scattering across the separatrix. The theory is extended to include the effect of this artificially enhanced scattering.

[FP1.122] Measurement of Trapped-Particle Mode Damping in Electron Plasmas.

Measurements of trapped-particle modes in pure electron plasmas have established the mode damping as a function of magnetic field and plasma temperature. The modes propagate on plasma columns when potential or magnetic field variations along the column cause local particle trapping. The modes consist of E \times B drifts of trapped particles, partially shielded by axial flows of untrapped particles.(A.A. Kabantsev et al.), Phys. Rev. Lett. 87, 225002 (2001). The damping occurs due to diffusion of particles across the velocity-space separatrix between trapped and untrapped fractions.(See poster by T.J. Hilsabeck and T.M. O'Neil, this session.) The measured damping rates are proportional to the number of particles in a boundary layer near the separatrix; exhibit an exponential decrease with temperature; and change magnetic scaling from B^-1 to B^-0.5 depending on rigidity. The damping is increased dramatically when a RF-drive is applied in resonance with the bounce motion of the marginally trapped particles, enhancing their diffusion across the separatrix. This diagnostic technique can ``map out'' the separatrix, or determine the distribution function along it.

[FP1.123] Resonant and Non-Resonant Thermal Fluctuations in Pure Electron Plasmas.

Spectra of thermal fluctuations in quiescent pure electron plasmas have now been measured at Trivelpiece-Gould modes and at non-resonant frequencies between modes. Each weakly damped mode has time-averaged electrostatic energy \textstyle \frac12 k_BT, where T is intermediate between the plasma temperature T_p and the receiver (load) temperature T_L. Experimentally, the spectra determine T_p and T_L as well as the internal and load-induced mode damping rates \gamma_p and \gamma_L. Away from a mode resonance, thermal fluctuations of N particles with energy N \cdot \textstyle \frac12 k_B T_p are predicted to be reduced by Debye shielding to a level \delta N \propto \sqrtN \; (\lambda_D / r_p )^3. The spectrum between modes is thus most informative at high plasma temperatures ( \lambda_D / r_p \geq 0.3 ) where the modes are strongly damped ( \gamma_p / ømega \geq 10^-2 ).

[FP1.124] Simulations of Damping of Trapped Particle Asymmetry Modes in Non-Neutral Plasma Columns

Kabantsev et al.(A. A. Kabantsev, C. F. Driscoll, T. J. Hilsabeck, T. M. O'Neil and J. H.Yu, in Non-Neutral Plasma Physics IV), AIP Conference Proceedings 606, 2001, pp. 277-286 have reported experimental observations and theory for trapped particle asymmetry modes on cylindrical electron columns. In particular, the m=1; k_z=odd mode exhibits strong damping from an unknown mechanism that is conjectured by Kabantsev et al. to be either diffusive mixing of trapped and untrapped populations of particles or spatial Landau damping. We have observed similar damping within a 3-dimensional particle-in-cell simulation. The simulation model does not include diffusive mixing. Spatial Landau damping is also ruled out because the mode frequencies in the simulation intersect the rotation frequency curve outside the plasma. We describe efforts to isolate the mechanism of the damping.

[FP1.125] A comparison of experimentally-determined plasma oscillation frequencies with simulation.

The frequencies of the various plasma oscillation modes (diocotron and Trivelpiece-Gould modes) in a non-neutral electron plasma are dependent on both the plasma temperature and on the plasma density profile. There have been difficulties in calculating mode frequencies that match those determined experimentally. We have experimentally measured the frequencies of the diocotron and T-G modes as well as the plasma density profile and temperature for several plasma conditions. These frequencies are compared with those obtained from kinetic simulations to determine the conditions necessary for a consistent fit to all the measured frequencies.

[FP1.126] An analytic model for the spectrum of the noise temperature diagnostic

The spectrum of image charge fluctuations under an isolated ring in a nonneutral plasma in a Malmberg-Penning trap carries information about the velocity distribution function of the plasma. We use this as a temperature diagnostic for our plasma. We previously used a particle simulation to derive the expected spectrum to compare with our experiment. However, this was time consuming and cumbersome.

We have recently derived an analytic model for the expected spectrum using randomly distributed noninteracting particles that bounce between the two ends of the plasma. The resulting power spectrum is a superposition of successively wider and smaller versions of the velocity distribution. The theory matches the results of the simulation with no adjustable parameters. This means that with an absolute calibration of our amplifiers we can not only derive the temperature from this diagnostic, but also the total number of particles in the plasma.

[FP1.127] Ion Trapping in the Virtual Cathode of PFX-I

The long-range goal of the Penning Fusion eXperiment-Ions (PFX-I) is the production of thermonuclear conditions by means of spatial and/or temporal compression of a high temperature, positive ion plasma. The present approach involves the confinement of positive ions in a virtual cathode produced by a non-thermal electron plasma held within a Penning trap of modified geometry. Spatial compression will be accomplished by shaping the virtual cathode to provide focusing of ion orbits, while temporal compression may be achieved by modulating the cathode strength and hence parametrically driving the ion motions. We will review data on the electron/ion plasma including the electron energy distribution and evidence for ions trapped in the virtual cathode. We will then present results from initial attempts to tailor the shape of the virtual cathode and to drive the ion motion by modulating the virtual cathode strength.

[FP1.128] The Columbia Non-neutral Torus and the Physics of non-neutral plasmas confined on magnetic surfaces.

The physics of non-neutral plasmas confined on magnetic surfaces is fundamentally different from that in previously studied configurations. We discuss the warm fluid equilibrium equation for a pure electron plasma [T. S. Pedersen and A. H. Boozer, PRL 88 (2002) p. 205002]. The equilibria are always minima of a suitably defined energy; however, this energy may not be the free energy of the system, so physical stability is not guaranteed by this theorem. We present 2-D calculations of equilibria for various boundary conditions, for both warm and cold plasmas and discuss implications for transport and stability. The basic physics of such plasmas can be addressed in the Columbia Non-neutral Torus (CNT), a table-top ultrahigh vacuum stellarator to be built at Columbia University. The CNT experiment is being designed to provide significant and variable magnetic shear, rotational transform, and magnetic field strength. We will discuss the physics basis of the CNT design and the basic physics capabilities of the experiment, including its possible future use as a positron-electron plasma trap.

[FP1.129] Status of the Columbia Non-neutral Torus

The Columbia Non-neutral Torus (CNT) is a tabletop (R=0.3 m, a=0.1 m, B=0.2 T) stellarator to be built at Columbia University. The goal of CNT is to study the equilibrium, stability, and transport of non-neutral plasmas confined on closed magnetic surfaces. CNT will use four circular coils, two interlocking coils with a variable tilt angle, plus two additional poloidal field coils located above and below the interlocking coils. By varying the angle between the interlocking coils, the configuration can be varied continuously from an aspect ratio of 2.4 with 33% shear to an aspect ratio of 4.3 with a shear of –7%. CNT will be designed to reach neutral pressures of 10^-10 Torr. The electrons will be injected from a multi-sectioned tungsten filament probe placed directly on the magnetic surfaces. The bias voltage and current flowing through each section of the filament can be varied to explore a variety of emission profiles. The plasma will be diagnosed by numerous Langmuir and sector probes, connected to a PCI-based data acquisition and control system.

[FP1.130] Thermal Fluctuations in Pure Electron Plasmas

Two methods have recently been described for determining the temperature of pure electron plasmas by measuring the thermal fluctuation spectrum on a surrounding sector probe. The first[1] employs the narrow resonant peaks associated with Trivelpiece-Gould modes of the the plasma, and the second[2] makes use of the broad continuous spectrum associated with independent particle motion. We propose a simple model, using the warm plasma dielectric function, for calculating the fluctuation spectrum. It is valid for both approaches and allows us to compare the two approaches. We find that there is a broad low frequency spectrum in the frequency range 0 < ohmega < kV, on which are superimposed peaks, one associated with each mode. V is the particle thermal velocity and k is the axial wave number. The frequency of a mode is closely given by cold plasma theory, but the height and width of a mode is the determined by Landau damping, as shown in [1]. The broadband spectrum associated with independent particle motion presented in [2] is not, in general, correct since the particles are not independent and correlation effects must be taken into account. The modes are seen to emerge from the continuum as their Landau damping decreases.

[1] Francois Anderegg, et al, CP606, Non-Neutral Plasma Physics IV, AIP 2002, p253. [2] M. Takeshi Nakata, et al, CP606, Non-Neutral Plasma Physics IV, AIP 2002, p 271.

[FP1.131] Experimental and Theoretical Studies of Electrostatic Confinement

Experimental and Theoretical Studies of Electrostatic Confinement J. Park, R. A. Nebel, C. P. Munson, W. G. Rellergert, M. D. Sekora Los Alamos National Laboratory Previous theoretical work [R. A. Nebel, D. C. Barnes, Fusion Technology (1998) and D. C. Barnes, R. A. Nebel, Phys. Plasmas (1998)] suggested that an ion cloud confined by a stable oscillating virtual cathode may undergo a self=similar collapse producing periodic and simultaneous attainment of high densities and temperatures. We are currently conducting experiments to test the stability of these virtual cathodes. Emissive probes have been used to measure time and space resolved potential and electron density profiles. Fluctuations in the plasma have been measured by a passive receiver and a combination of an external driver and a receiver. The observed virtual cathode exhibits a bifurcation between states where the well depth is ~ 60potential. The transition is a function of the injected electron flux, grid biases, and the gas pressure. Experimental results on fluctuation and stability of a driven virtual cathode will be presented and compared with theoretical predictions [R. A. Nebel, J. M. Finn, Phys. Plasmas (2001)].

[FP1.132] Flat-top conundrum

A truism of non-neutral plasma physics research is that flat-top (i.e. n(r)=\hboxconst for r

[FP1.133] Ion Resonance Instability in a Double Well Trap

Double well traps intended for the production of anti-hydrogen must have regions of overlap between the positrons and anti-protons. This overlap opens the possibility of instabilities, and is similar to the contamination of a single species column by a small number of particles of opposite sign. In 1969, Levy, Daugherty and Buneman predicted that such contamination could lead to a surface instability, whch they named the ion resonance instability. Previously, the l=1 instability had been detected. New measurements detect the l=2 and l=3 instability. More importantly, the instability occurs in a regime entirely different than that predicted by Levy et al. Their instability was driven by shears created by mass differences; here the instability is driven by shears created by the difference in confinement regions for the ions and electrons, and the different electric fields therein.

[FP1.134] Electron Injection into Malberg-Penning Plasma Traps

The process of continuous electron injection into plasma traps is investigated experimentally and with Particle-in-Cell (PIC) simulations. Temporal measurements of the trapped electron population and axially averaged transverse density show unexpectedly rich dynamics. Similar features are observed in PIC simulations. The number of electrons in the trap increases over hundreds of axial bounce times and exhibits distinct regimes of growth. The population increases beyond that realized through 'ballistic' filling (which requires just two bounce times). This increase is initiated by the development of a two-stream interaction between electrons emitted from the cathode and reflected electrons from the end-plug of the trap. PIC simulations show that the axial phase space distribution develops in a remarkable way: a series of "bubbles" appear, which, as they oscillate, redistribute the electrons in phase-space, increase the number of trapped electrons and lead to a more uniform phase space density. These bubbles have been observed experimentally. Very long injection times exhibit signs of radial potential matching of the plasma to the emitting cathode.

[FP1.135] Buffer gas cooling of pure-electron plasmas

We will describe efforts to cool pure-electron plasmas using inelastic collisions with a room temperature buffer gas. In the initial experiments, a 1 eV gas of electrons confined in a 1.5 kG Malmberg-Penning trap will be cooled using carbon dioxide. CO2 was chosen for initial testing because of its large ratio of inelastic to elastic scattering cross-sections in the 1 eV range. If successful, this cooling method could be used for rotating electric well confinement techniques at low axial magnetic fields. It will also be used to investigate rotating magnetic quadrupole techniques.

[FP1.136] Three-dimensional PIC codes for non-neutral plasmas.

With the recent advances in computer speed and memory, three-dimensional (3d) computer codes have become practical or nearly practical on single user workstations. Two-dimensional (2d) codes have already proved very useful in understanding problems in non-neutral plasmas. We will report on progress towards an efficient 3d code tailored towards problems in non-neutral plasmas. The electrostatic particle-in-cell (PIC) code will use cylindrical coordinates to match the typical trap boundaries, and we will begin by implementing E\timesB dynamics perpendicular to the external magnetic field and full dynamics along the magnetic field. The code will be based on the 2d codes developed by the Plasma Theory and Simulation Group at Berkeley (see Verboncoeur et al., J. Comput. Phys. 104 (1993)). Once in operation, we will used the codes to explore problems such as non-axisymmetric trapped two-stream instabilities, and quadrupole induced resonant particle transport.

[FP1.137] Nonlinear PIC Simulations for Nonneutral Plasmas

We present nonlinear simulations of the low frequency dynamics of electrons in a Malmberg-Penning trap, including compressional and thermal effects [1,2].

First, we consider a 2D model where we assume the effective plasma length constant in time. In this framework, we further neglect the thermal effect on the velocity field, and show with the PIC code KANDINSKY that Penning traps could be used to perform geophysical fluid dynamics experiments [3]. We also observe that, due to the presence of the nonlinear m=1 instability, the initially hollow density profile becomes peaked, as in the experiments.

Then, we show 2D results including thermal effects. In this case, the development of the m=1 instability is slowed since the equilibrium plasma length profile is closer to the integrable profile, namely the length profile for which there are no discrete unstable modes [4].

Finally, we present simulations of the 3D fluiddynamics model of Ref. [2]. In particular, we investigate the evolution of a m=1 perturbation for different electron temperatures, when compressional and thermal effects are included.

[1] J.M. Finn, D. del-Castillo-Negrete, D.C. Barnes,\textitPhys. Plasmas, \textbf6, 3744, 1999.

[2] G.G.M. Coppa, A. D'Angola, G.L. Delzanno, G. Lapenta, \textitPhys. Plasmas, \textbf8, 1133, 2001.

[3] G.L. Delzanno, J.M. Finn, G. Lapenta, "Nonlinear Phase of the Compressional m=1 Diocotron Instability: Saturation and Analogy with Geophysical Fluid Dynamics", submitted to \textitPhys. Plasmas.

[4] G.L. Delzanno, V.I. Pariev, J.M. Finn, G. Lapenta, "Stability Analysis of Hollow Electron Columns Including Compression and Thermal Effects: Integrability Condition and Numerical Simulations", submitted to \textitPhys. Plasmas.

[FP1.138] Electron diffusion in the annular Penning trap

Transport by cross-field diffusion has been studied in the annular Penning trap in which a nonneutral plasma of electrons is contained between concentric cylinders. At densities sufficiently low (<10^5 cm^-3) to suppress mobility transport arising from the space charge electric field, the dominant sources of transport are diffusion from collisions of electrons with added helium gas and asymmetry transport from stray fields. The collisional diffusivity is shown to scale linearly with collision frequency and inversely with the square of the axial magnetic field. The measured mean energy is initially 0.3 eV and the least energetic electrons are lost more slowly as a consequence of the energy dependence of the diffusivity. Decay constants are about a factor of four higher than calculated from the electron-helium momentum transfer collision frequency. Both the asymmetry transport and the collisional transport are shown to depend upon the cleanliness of the trap surfaces.

[FP1.139] Limitations on Confinement of a Toroidal Electron Plasma due to Field Asymmetries and the Presence of Neutrals

Issues of equilibrium, stability, and limitations on confinement for toroidal electron plasmas are experimentally addressed using the Lawrence Nonneutral Torus(M.R. Stoneking \textitet al.,) Phys. Plasmas \textbf9, 766 (2002).. Electron densities in the range of 10^6 cm^-3 are trapped for up to 300 \mu s in a partially toroidal trap (B=200G, R_o=43cm, a=5cm). A horizontal electric field is required to provide equilibrium force balance in the major radial direction. In this paper we discuss experimental studies of the ion resonance instability (due to ionization of background neutral gas) and its effects on confinement. We assess the effects of field asymmetry on confinement by independently applying a vertical correction field and a field error. Initial results using a new phosphor screen imaging detector are presented. In addition, designs for an upgrade to the experiment are presented. The upgrade will enhance the magnetic field by \approx 5 times (to \approx 1 kG), improve vacuum conditions (to 10^-9 Torr or better), improve field symmetry and lower the aspect ratio. This work is supported by U.S. Dept. of Energy and Lawrence University.

[FP1.140] Finite length diocotron mode in ELTRAP

Electrostatic perturbations of the electron plasma in the Malmberg-Penning trap ELTRAP have been investigated using the cylindrical electrodes as probes, and a low noise differential amplifier. The FFT exhibits a very high peak on a noise spectrum. This frequency corresponds to a m=1 diocotron mode and remains constant for a time interval which depends on the magnetic field, then decreases. The variation is related to particle losses: at lower residual gas pressure and higher B fields the oscillation is more and more persistent. The amplitude of the oscillations grows as long as the frequency remains constant, then saturates (corresponding to plasma touching the walls) and decreases with frequency. The mode amplitude grows in time following a power law at low magnetic field (B<300 G) and exponentially at higher fields. It is found that the excitation of the mode occurs at the start of the hold phase of the cycle, when the potential barrier is raised to confine electrons. Measurements performed with increasing rise time have shown that the growth rate decreases and at long enough rise time the instability does not arise.

[FP1.141] Longitudinal cooling of a strongly magnetized electron plasma

The optimal values of Q and \Delta, the detuning of a microwave frequency from resonance, for cooling a pure electron plasma with a microwave bath have been calculated. An electron plasma which has no internal degree of freedom, cannot be cooled down below a heat bath temperature. However, the longitudinal cooling can be achieved by energy transfer from the poorly cooled parallel degree of freedom to the well cooled perpendicular degree of freedom. A microwave tuned to a frequency below the gyrofrequency of the electron forces an electron moving towards the microwave to absorb a photon and then to move up one in Landau state. The electron loses longitudinal energy in this process. On the basis that most of the electrons are in the ground or first excited state, we set up a transition equation and develop a FEM code. With an appropriate condition for B-field and intensity of the microwave, the cooling times for several values of Q's and \Delta's are calculated and the optimal values are found. Applying the optimal values at appropriate times in a cooling process, the best cooling can be obtained. For an electron plasma magnetized with 10T B-field, cooling to the solid state can occur within 2 hours.

[FP1.142] ICF Technologies

[FP1.143] Ultrafast X-ray Difraction for Measurements of Structural Dynamics in Shocked Metals

An experiment on structural dynamics at the ultra-fast time scale in shocked metal samples is presented. The technique development of an ultrafast x-ray diffractometer to generate "molecular movies" is described. Preliminary results of static x-ray measurements of thin unshocked Ga samples are presented. Initial experiments use 200-300 mJ of a 100fs Ti:Sapphire laser to excite K-alpha x-ray emission in an aluminum wire. The x-ray emission is relayed using a spherical crystal to the sample target. Plans for experiments using Cu K-alpha emission will also be described.

[FP1.144] Two-dimensional ablation density measurement with micrometer resolution Fresnel phase zone plate

The ablation density is a critical quantity for good understanding and stabilizing the Rayleigh-Taylor instability. However, ablation density measurement has never been performed. High spatial resolution (<3 \mum) hard x-ray (\sim5 keV) imaging technique is necessary to measure the ablation density. We have developed Fresnel phase zone plate (FPZP) imaging technique, which can image hard x rays with high spatial resolution. Using 4.7 keV x rays, we have obtained the FPZP spatial resolution of 2.2 \mum. We performed the ablation density measurement experiment of the laser irradiated polystyrene target. In this experiment, we coupled the flash backlight technique with the FPZP imaging, and obtained high spatial and temporal (\sim150 ps) resolution. We obtained density profiles of the in-flight polystyrene target at different three timings. We compared the measured density profile to the prediction of the 1-D hydrodynamic simulation code.

[FP1.145] Kinetics calculations for K-shell Kr multi-keV x-ray production experiments at OMEGA

Efficient, multi-keV x-ray sources are necessary for radiography of high-density ICF fuel capsules. We present calculations for the emission of K-shell Kr radiation (h\nu\geq~13keV) from high-temperature, high-density plasmas. The models include fine-structure energy levels and complete manifolds of dielectronic recombination channels. The collisional-radiative emission for transitions with energies above 13~keV is computed with radiative transfer through a 50~\mum plasma with the CRETIN code. Account is taken of plasma temperature and density gradients as predicted from LASNEX hydrodynamic simulations. Stark broadening for the emitted line profiles is computed, as well as source broadening and the instruemental resolution effects. The observed total Kr K-shell radiative yield is compared to both XSN-based and recent, detailed predictions. The Kr K-shell is seen to be an efficient, high-photon-energy radiator in recent experiments at the OMEGA laser facility.

[FP1.146] Non-LTE Kinetics modeling of krypton ions : calculations of radiative cooling coefficients

Radiative yields from high-Z ions are important in understanding energy distributions and spectral characteristics in plasmas. The understanding of x-ray yields is important to the development of laboratory radiation sources. Since these plasmas occur over a wide range of plasma conditions, they requires a general non-LTE population kinetics description. We investigate radiative properties of non-LTE krypton plasmas with a collisional-radiative (CR) model constructed with more than 80,000 levels covering more than 12 ionization stages. We present the first detailed calculations of total radiative cooling coefficients and the x-ray radiative cooling coefficients of krypton ions as a function of electron density above the coronal limit. Special attention has been given to the process of dielectronic recombination. Averaging schemes for level structures and configuration populations have been investigated in order to produce a more computationally tractable model. Radiative cooling coefficients are given for plasma conditions from 600 eV \leq T_e \leq 10~keV and 1\times10^14 \leq N_e (cm^-3) \leq 1\times10^24. Ionic radiative cooling coefficients as well as steady-state calculations of the average charge state at given plasma conditions are also presented.

[FP1.147] An Anomaly in the Inglis-Teller Limits of the C VI Lyman and Balmer Series in Laser-Produced Plasmas

Soft x-ray spectra from thin carbon layers heated by the OMEGA and NIKE lasers have been obtained with both spherical and planar targets, respectively, using a flat-field grazing incidence spectrograph equipped with a gated microchannel plate for temporal resolution. In both experiments, late-time (recombining) hydrogenic C VI spectra show an n-to-1 Lyman spectral series blending with the continuum at n=4, contrary to n=9 in the n-to-2 Balmer series. It appears unlikely that plasma inhomogeneities are the sole cause of this anomaly, given the difference in the experimental configurations. Other explanations for the line-to-continuum merging (other than the usual Stark-broadened Inglis-Teller effect) under consideration include non-thermal Doppler broadening, deviations from statistical sublevel population distributions, and opacity effects. Collisional-radiative and hydrodynamic modeling, including cascades, is employed to further understand this phenomenon.

[FP1.148] Modeling of x-ray detection using microchannel plates

X-ray diagnostics involving microchannel plates (MCPs) are used for imaging and spectroscopy in laboratory simulations of astrophysical systems and in many other types of experiments. Modeling of the MCP and measurements of MCP performance can be useful in optimizing these measurements. Some early detailed modeling of MCP performance [Fraser, Nucl.Instr. and Methods, 195 (1982)] was focused on pulse counting applications. Present-day measurements often involve the measurement of amplified signals; we discuss modeling of this case. We focus specifically on the effect on the signal of the distribution of depths and angles at which the x-ray photons strike the channel walls. In addition, we will show results from and the status of the facility we have built at Michigan for testing MCPs and x-ray framing cameras.

[FP1.149] Spatial structure data of ultra-thin Au films for instability modeling

Non-uniformities in a thin high Z overcoat such as Au are a potential contributor to the seed for Rayleigh-Taylor instability.There are insufficient data on the spatial structure distribution of thin Au film overcoats used in the instability experiments for modeling. This may contribute to the discrepancies observed between the modeling and the experimental results where Au films with thickness less than 100 angstroms were used. In this presentation, we will provide AFM analysis of thin Au films with averaged thickness of 50 to 200 angstroms. Digital lineouts will be included in the data set to be used directly for modeling.

[FP1.150] Planar Target Fabrication in Support of HEDP Research

We will present an overview of techniques used to fabricate and characterize targets used in hydrodynamics and equation of state experiments for HEDP research. Planar targets made at Schafer Laboratories are used for experiments on NRL’s Nike KrF laser, the Omega laser at The University of Rochester’s Laboratory for Laser Energetics, and other drivers. Recent experiments have demanded increasing target complexity. Targets for EOS experiments may have several layers of material including Kapton, micromachined aluminum flats or steps, and witness plates consisting of glass supports, thin polyimide covers, and aluminum stripes. Targets for hydrodynamics experiments consist of various densities of material with patterned surfaces, and metallic coatings. We will present our present capabilities for these kinds of targets and what we foresee as the next generation of plastic and foam targets. This work is supported by the U.S. Department of Energy under contract DE-AC03-01SF22260.

[FP1.151] Modeling of indirect laser-driven ICF implosions and experimentally observable spectral properties.

K-shell emission spectroscopy is commonly used to diagnose core temperature and density of Ar-doped ICF implosions at OMEGA and Z. To investigate details of spectra formation, we perform simulations of Ar-doped indirect laser-driven implosions and generate synthetic spectra and plasma core images observable at the collapse of the implosion. VisRad is a user-friendly view factor code used to simulate the radiation environment in three-dimensional objects. It predicts temperature and radiative flux distributions throughout target components in high-power laser and z-pinch laboratory plasma experiments. Those time-dependent hohlraum radiation temperatures and fluxes are used to initialize target implosion simulation with a 1D rad-hydro code BUCKY. Plasma core temperature and density distributions from the hydro code are then used to compute Ar emission spectra and core images. To this end we utilize multi-dimensional collisional-radiative, spectral analysis code SPECT3D. We will discuss details of the calculations and compare our results against experimental data.

[FP1.152] Effects of Line Shifts and the Ion Quadrupole Contribution on Spectral Line Assymetries

With the incorporation of full Coulomb electron broadening models into our Stark spectral line broadening code MERL, we have been able to successfully predict experimentally observed spectral line shifts in K- and L-shell argon spectra emitted from ICF microballoon implosions reaching electron densities on the order of 1-3\times 10^24/cm^3. We have also looked at the possible effects of these shifts on spectral line merging as electron density increases. We will now focus on asymmetries in these line spectra with a concentration on the line wings. The two full Coulomb electron- broadening models used in our line spectra code are currently predicting different amounts of red/blue wing asymmetry that we will scrutinize in greater detail. We have also successfully incorporated the ion quadrupole effect into the calculation of these spectra and will look at its effect on this asymmetry as well as its effect on the merging of these line spectra as electron density increases. Finally, we will use these theoretical spectra to analyze experimental data from upcoming ICF implosion shots on the OMEGA laser system of microballoons containing deuterium and small amounts of argon.

[FP1.153] Improving Numerical Efficiency in Three-dimensional ICF Target Simulations

This paper describes some improvements to the numerical efficiency of the three-dimensional radiation hydrodynamics code FastRad3D. FastRad3D is a compressible hydrodynamics code containing most of the physical effects relevant for the simulation of high-temperature plasmas including inertial confinement fusion (ICF)-regime Rayleigh-Taylor unstable direct drive laser targets. These effects include inverse bremmstrahlung laser energy absorption, classical flux-limited Spitzer thermal conduction, real (table look-up) equation-of-state with either separate or identical electron and ion temperatures, multi-group variable Eddington radiation transport, and multi-group alpha particle transport and thermonuclear burn. FastRad3D uses an MPI message-passing model to obtain parallelism on workstation clusters and supercomputers. We discuss improvements to the temperature solver and to the transport algorithm during the burn phase of the calculation. The latter employs the BIC algorithm treating the pressures implicitly and other transport variables explicitly to eliminate the sound speed limitations on the time step size. A sample calculation with timings on various computer architectures will be shown

[FP1.154] The VNIIEF/LANL Collaboration: Ten years of scientific benefit to the Russian Federation and the United States

Since 1992, the All-Russian Scientific Research Institute of Experimental Physics (VNIIEF) and the Los Alamos National Laboratory (LANL), the institutes that designed the first nuclear weapons of the Soviet Union and the United States, respectively, have been working together in fundamental research related to pulsed power and high energy density science. Experimental and theoretical work has been performed at VNIIEF and LANL in areas as diverse as imploding z-pinch liner physics and applications, fusion plasma formation, isentropic compression of noble gases, and explosively driven high current generation technology [1]. Recent joint work has focused on the Atlas capacitor bank (23 MJ, 30 MA, 6 microsec) now operational at LANL. Even before Atlas became operational, VNIIEF's magnetic flux compression capability was used to provide the US with the first available data at ATLAS' upper performance limit (31 MA, 4 microsec, 12 km/s velocity for 50 g liner mass). VNIIEF has recently designed and fielded imploding liner experiments on Atlas, with the goal of studying material strength properties by observing instability growth. The collaboration is reviewed and new results are reported.

[1] "US/Russian Collaboration in High-Energy-Density Physics Using High-Explosive Pulsed Power: Ultrahigh Current Generation, Ultrahigh Magnetic Field Applications, and Progress Towards Controlled Thermonuclear Fusion," I. Lindemuth et al., IEEE Trans. Plas. Sci. 25, 1357 (1997).

[FP1.155] Energy Transport in a Wall-Confined Inverse Z-Pinch

Direct confinement of high-beta plasma by material walls appears an attractive option for fusion (MTF) but has been studied only a little. The well-benchmarked 2-D rad-MHD code MHRDR is being used to design an inverse z-pinch experiment, driven by the 2-TW Zebra generator, to study MTF transport and confinement. According to MHRDR, the plasma is expected to evolve into a near-equilibrium, with thin wall sheaths that contain steep temperature and density gradients. The plasma should take about 3 microseconds to cool, even in the presence of considerable convection. This is much longer than if free-streaming losses of ions or unmagnetized-electron conduction losses were present. This would make MTF attractive, if borne out by experiment. A parametric study of the inverse pinch, as a function of chamber geometry, chamber dimensions, seed magnetic field, and initial gas pressure, is helping to design the experiment. Modeling has found a region of parameter space with adequate heating, formation of a quasi-static magnetic equilibrium, and near-classical cooling rate, even in presence of substantial plasma convection. Experimental verification of the energy transport in this simple wall-confined plasma would increase confidence in the design of integrated liner-on-plasma experiments.

[FP1.156] Near-Field Nonuniformities in Angularly Multiplexed KrF Lasers

Induced Spatial Incoherence (ISI) is an effective technique for achieving the high degree of spatial illumination uniformity required for direct-drive fusion. Although ISI provides ultrasmooth illumination at the far-field of the laser, where the target is located, it may still allow the beams in the quasi near-field to develop significant time-averaged spatial nonuniformities. This structure, which arises primarily from random phase distortion and Fresnel diffraction, increases as the beams propagate away from the pupil plane images located at the amplifiers; it is distinct from any structure imposed by amplifier gain nonuniformities. Because of the spatial incoherence of ISI beams, the time-integrated structure is significantly smaller than that experienced by coherent beams. Nevertheless, it remains a potential optical damage issue, especially in the long delay paths required for large angularly-multiplexed KrF lasers. This presentation compares simulations and measurements of quasi near-field structure in the Nike KrF laser, and presents simulations showing how optical relaying can control the problem in a future KrF driver.

[FP1.157] Amplifier Modeling and Scaling for the Electra krF Laser

The rep-rated KrF laser system Electra is one component of the High Average Power Laser Program. A simulation code, Orestes, has been developed to support experiments on the Electra gas laser amplifier, to improve efficiency, and to scale to larger systems such as the Integrated Research Experiment (IRE). Orestes is a first principles physics code which includes e-beam ionization and excitation, plasma chemical kinetics with 1-D spatial resolution along the lasing axis, detailed vibrational structure of the KrF molecule, transport of lasing photons, and time-dependent transport of amplified spontaneous emission in 3-D with line narrowing. Validation of the code is based on comparison with high powered laser experiments. Predictions for the laser output from Electra both as an amplifier and an oscillator are presented as a function of total pressure and composition. These figures indicate the impact of 3-body relaxation on the predicted laser output and specify the optimal conditions. Initial studies of the configuration and yield for the IRE will also be discussed.

[FP1.158] Efficient Energy Deposition for an Electron Beam Pumped KrF Laser

Electra is a repetitively pulsed, electron beam pumped krypton fluoride (KrF) laser that will develop the technologies that can meet the Inertial Fusion Energy (IFE) requirements for durability, efficiency, and cost. The Electra laser is pumped with two opposing electron beams each with parameters of 500 kV, 90 kA, with a 100 ns flat-top pulse duration, and a cathode area of 27 x 97 cm^2. The e-beam propagates through a hibachi structure, which supports a thin foil that isolates the vacuum diode from the high-pressure (>1 atm) laser gas. It has been demonstrated that segmenting the beam into strips to miss the hibachi support ribs significantly increases the electron beam deposition efficiency. The energy deposition efficiency is defined as the ratio of energy deposited in the laser gas over the vacuum diode e-beam energy. Energy deposition efficiencies of 75have been achieved with a 500 keV e-beam. In addition, 1-D and 3-D codes have simulated the e-beam propagation through the hibachi, and 1-D codes predict a maximum energy deposition efficiency of 81

[FP1.159] Space-variant phase/polarization of ultra-short-pulse laser, I - Generation of intense optical vortex and intense longitudinal electric field

The common knowledge of Hermite/Laguerre mode conversion can introduce a new aspect into the research of ultra-short-pulse laser plasma interaction. Among several methods for the mode conversion, we adopted a spiral phase plate with the first order topological (azimuthal) charge that generates not only an intense optical vortex but also the first order Bessel beam with using a phase element like an axicon lens. In addition the conversion of polarization state is also interesting. A beam with radial polarization has a longitudinal electric field near a focus. We developed a liquid crystal device that uses the space-variant optical rotatory. This device can efficiently generate the axially symmetric (radial and azimuthal) polarization from the usual linear polarization. We will report in detail the fabrication processes of these optical elements, phase and polarization properties of mode-converted beams and the measured longitudinal electric field distribution.

[FP1.160] Space-variant phase/polarization of ultra-short-pulse laser, II - Preliminary observation of interaction with plasma

In the laser-plasma interaction, the motion of electron depends on spatial variations of the polarization and the phase of incident laser beam. We developed special optical elements (a spiral phase plate and a liquid crystal polarization rotator) for generating the axially symmetric (radial or azimuthal) polarization and the optical vortex (Laguerre-Gaussian beam) with linear or circular polarization. These optics were installed into a Ti:sapphire laser system (T^6 laser) that deliver ~100mJ/100fs at 800-nm wavelength. We will present density profile, induced magnetic field (both measured using UV probe beam synchronized to an interaction beam), energy spectra of forward accelerated electrons and their dependence on the spatial variation of phase and/or polarization.

[FP1.161] Calculation of the Response of Inertial Fusion Energy Materials to X-ray and Ion Irradiation on Z and RHEPP

The response of candidate Inertial Fusion Energy first wall materials to the intense bursts of x-rays and ions that will emanate from IFE targets is an important issue for power plants and for any facility that would test high yield target performance. In an effort to gain understanding of the behavior of these materials while undergoing rapid phase changes and to validate code that predict such behaviors, a multi-institution group has been conducting experiments on the Z and RHEPP facilities at Sandia. As part of this project, calculations of irradiation with tungsten wire array x-ray on Z and ions on RHEPP have been performed with the BUCKY computer code. When comparing experiments to calculations for graphite and tungsten samples, we have found some behavior that is not predicted by the code. We will show how the greater than 1 keV lines in the Z tungsten wire array x-ray source affect the material response. On the RHEPP ion experiments, we will show how sensitive the results are to the thermal conductivity of the material.

[FP1.162] IFE Chamber Wall Materials Response to Pulsed X-rays on Z and Ions on RHEPP* **

We are investigating the response of candidate Inertial Fusion Energy (IFE) reactor materials to pulsed x-rays (on Z) and energetic ions (on RHEPP) at fluences similar to those predicted for future reactors. Both the Z and RHEPP facilities are located at Sandia National Laboratories. Primary dry wall armor materials under consideration are carbon and tungsten in various forms, in either flat or 'engineered' geometry such as carbon 'velvet'. Experiments are underway to 1) determine single-shot threshold for ablation, 2) characterize surface roughening occurring below this ablation threshold, and 3) measure net ablation for doses exceeding the ablation threshold, as a function of dose. Graphite is observed to ablate readily above a threshold dose level of 3-4 J/cm^2. Sintered graphite and tungsten sheet are observed to exhibit thermomechanical responses that result in material loss beyond that expected from sublimation predicted by BUCKY and other modeling codes. Experiments are planned to measure erosion of the carbon velvet tips and shafts due to RHEPP ion exposure, to measure directly the time-dependent surface temperature during ion beam exposure, and to compare both with code predictions.

[FP1.163] Response of Dry Wall Chamber Designs to the Output Spectrum from a Directly Driven Laser IFE Target

One dimensional radiative hydrodynamic simulations of the response of gas protected dry wall chambers to the output of high gain laser direct drive targets are presented. It is found that the dominant threat is from the ionic component of the target output. Variations in chamber operating temperature, radius, target yield and buffer gas composition are considered, and a variety of options satisfying both the constraints of per-shot chamber wall degradation and target injection are presented. Satisfying these constraints is necessary, but nor sufficient to assure a successful chamber design. The amount of implantation of high energy burn products into the wall for each of the variations consiered will be discussed.

[FP1.164] On Afterglow Plasma Parameter Evolution and Heat Flux to the Target*

Afterglow stage of plasma/gas temporal evolution in the ife chamber plays very important role both for protection of the chamber first wall from plasma impact damage and for favorable conditions of next pellet injection which will follow in a time scale ~0.1s [1]. In this report we analyze main physics processes leading to residual gas/plasma cooling and plasma recombination. We develop a time dependent numerical model to study the effects of large convective cells, plasma/gas radiation opacity on cooling and recombination processes. We discuss the impact of residual plasma and radiation field on the heat flux to the target. [1] S. I. Krasheninnikov, "IFE Dry Wall Chamber Physics Issues", 29th EPS Conference on Plasma Physics and Controlled Fusion, 17-21 June 2002, Montreux, Switzerland

[FP1.165] Investigation of plasma development in magnetically insulated transmission lines

Plasma forming from metal surfaces in high power pulsed vacuum discharges limits the power transmitted to a load through magnetically insulated transmission lines, and eventually shunts the load, producing so-called MITL closure. An experimental investigation of the plasma formation and evolution is being developed with the goal of understanding and controlling the explosive electron emission, the plasma closure, which depend on the electrode conditioning and on the electric and magnetic field strength. Cylindrical coaxial electrode systems are driven by the 2 MV, 2 ohm Zebra generator, that produces load currents rising to 1 MA in 100 ns. The current flowing through the system is measured in detail with B?dot probes located before and after the MITL section. Faraday cups and radiochromic film are used to measure electron beams. To characterize the plasma produced in the anode-cathode gap, time-resolved imaging and spectroscopy, and laser diagnostics are being developed. The electrode initial conditions are carefully characterized and controlled. Preliminary experiments with a short circuit load showed that the plasma closure can be predictably controlled by changes in geometry, via the electric and magnetic fields. Computer simulations of the plasma formation and evolution in these experiments will be performed.

[FP1.166] Consideration on Cathode Structure in Mather-type Plasma Focus Devices

The hot, dense plasma produced in plasma focus is a rich source of phenomena such as the emission of intense radiation, neutron yield, as well as copious nuclear fusion products. One of the most important characteristics of the plasma focus is intensive neutron emission that is produced when deuterium is used as filling gas. There are a lot of experimental parameters required for optimizing neutron yield, such as length of center electrode, radii of electrodes, gas pressure, impurities, insulator material, geometry and so on. However, the optimum condition for each device is roughly speaking, obtained by pinching the plasma in front of the anode at the phase near the current maximum. When a strong focus occurs, the impedance of the pinch region increases dramatically, and the circuit energy is concentrated there. In order to increase the input energy on the pinch phase, the energy dissipation and energy loss in run-down phase of discharge should be small. In this paper, the influence of the cathode structure on production of neutron yield in a 7 kJ Mather-type plasma focus device have been investigated by using two types of cathode electrode, 1) bar and 2) tubular cathodes. In the present experimental results that were obtained under the same conditions except for the cathode structure, the energy dissipation in the run-down phase was much lower in the bar cathode than in the tubular one and as a result, the neutron yield was higher in the bar cathode. The experimental results obtained by many authors using both cathodes do not seem to support our experimental results. However, they were obtained in different devices with the bar or tubular cathode. The present result clearly shows that for a complete comparison it is necessary to investigate both types of cathode in the same device under the same optimum conditions

[FP1.167] Study of the neutron pulse regularity of a small dense plasma focus device by time of flight measurements

Plasma foci are well know to be efficient neutron generators, with a great potential for immediate applications. For this purpose, it is particularly important to assess the uniformity of their behaviour, which strongly depends on the geometry details and the materials used in their construction. It has been established that there are competing mechanisms in the neutron generation, that yield both isotropic and anisotropic components. However, the latter, which may be associated to the generation of axial ion-beams has been found to contribute no more than 30the total neutron yield, and less than 10particular device. As a matter of fact, measurements of the ratio of head-on and side-on neutron fluxes have been found to be grossly misleading, since they may be extremely large due to very pronounced anisotropic components, even though they may actually contribute less than 10neutron yield. The isotropic component, on the other hand, can be reasonably described by thermonuclear models, and within the experimental error, allowed by the regularity of the device behaviour, can be properly characterised by side-on measurements. The purpose of this work is to study the regularity of the neutron emissions from a small dense plasma focus (5kJ at 37kV) [1], analysing the time of flight signals from three, side-on, on-line, scintillation detectors.

[1] F. Castillo, J.J.E. Herrera, J. Rangel, A. Alfaro, M.A. Maza, V. Sakaguchi, G. Espinosa and J.I. Golzarri, "Neutron Anisotropy and X-ray Production of the FN-II Dense Plasma Focus", Brazilian J. Phys. Vol.32 (2002) 3-12.

Part F of program listing